Getting power from a USB port...

Hi,

I work on the train during my commute, and I'd like to be able to power my
uP development board from my laptop (its a small dev board), so I can do
code development and testing.

I thought about using the USB port to provide 5V, and then using a MAX1811
or LTC4053 to charge a 4.2V Li battery, which in turn feeds a LT1308B or
MAX1771 boost converter to give 5/9/12V etc. at a few 10s of mA.

Is this a reasonable idea? It shouldn't load the USB port too much - both
the MAX & LT Li chargers are designed to run from a USB port, so shouldn't
overload it, and the boost converter should give me fairly clean power at
whatever voltage I need. I can charge the unit at work from the desk PCs,
and therefore the loading on the laptop should be minimal for the hour on
the train.

Are there better ways of doing this? Is there a cheap commercial alternative
already available? This should end up very small and should cost about USD
15 to 20 when made on a PCB. Alternative chips would be a help too...

Thanks for your input,

Mike

 

I remember seeing some USB current limiter/ management chips, prolly
on maxim's site, might be worth search. There are some USB specs
saying you MUST NOT exceeds xxx milliamps. This is what the managemant
chips do, IIRC.

Wonder when they will bring out a USB soldering iron?


martin

After the first death, there is no other.
(Dylan Thomas)

 

There are two current levels available from USB ports. 100mA from Bus
Powered ports and 500mA from self-powered ports. I would assume a laptop
host port is self-powered. BTW, Manufacturers would only have to adhere
to these standards if they are planning on USB org certification.

 

I don't know that I'd count on a laptop USB port meeting specs written
for tower boxes.

...Jim Thompson

 

Ideally, (as far as the USB guys are concerned), there's individual port
overcurrent protection that is set to the maximum of 10mA (I think -- number
could be wrong) that the device is allowed prior to enumeration, and then set
to the "high end" power limit of the port (100mA or 500mA or whatever) once
enumeratoin is complete and the OS has determined that that much power
actually is availble. Several manufacturers have chips out there that provide
"per port" protection and have inputs for configuring whether the current
limit is set "low" or "high" (e.g., 10mA or 500mA -- the actual values are set
by resistors).

In actuality, it's rather _un_common to have a system that doesn't simply use
a "one size fits all" approach to current limiting: If there are, say, 4 USB
ports, they current limit the "+5V USB" line to 500mA*4=2A. The end. No port
by port protection, no low/high power modes, etc.! In effect, you peripheral
is connecting to a shared +5V bus that also an active fuse -- and nothing
else.

I'd be surprised if a laptop can't manage to put out 500mA per USB port (if
not more based on the above!) -- there are many "high power" (500mA) devices
out there, and the manufacturers don't want to get the phone calls from
everyone whose favorite peripheral doesn't work. That being said, though, I
onve used a KVM that was very sensitive to the USB voltage -- in theory it
should have worked down to something like +4.25V, in practive it appeared to
need more like 4.75V or above to function correctly!

---Joel

 

If you're going to use a battery anyway, why even worry about charging
it from the USB port? A couple of AA size NiMH cells will give you
around 1500 mAh at about 2.4 V, which is way more than you need. You
can charge them from AC at work and at home, and maybe even keep a
reserve set of AA alkalines in the laptop bag in case they run out
unexpectedly.

Having said that... I think you can get reasonable amounts of juice out
of the USB port. I was given a small "USB" aquarium that I have hooked
up to my computer at work. It has a blue LED and a small pump that
makes two plastic fish swim around. The power for the LED and pump can
come from either the USB port or some AA batteries, and this is selected
by slide switches on top of the tank. I don't think it touches the USB
data lines at all; I don't think it has a chip to do the full USB
negotiation and I don't see an aquarium show up in the device list when
I plug it in. I figure maybe 3.6 V at 20 mA for the LED and at least
that much for the motor.

If your laptop dates from the transition period when they still had PS/2
mouse/keyboard ports, you can also get 5 V from that port. Not sure of
the available current but it should be a few tens of milliamps at least.

http://www.tkk.fi/Misc/Electronics/circuits/lptpower.html says that if
you have an IEEE 1284 level II parallel port, you ought to be able to
get at least 2.4 V at at least 14 mA from each data pin.

There is probably a reasonable amount of juice available in the PCMCIA
card slots. You might even be able to get some of the circuitry onto
the card, further reducing the amount of stuff in an external box.

How about plugging into the sound card output and always having a loud
MP3 playing? Or a solar cell and some tape to stick it against the
light fixture on the train...

Matt Roberds