Lead acid life
--------------
Melton's recent post[1] on (among other important and interesting things) the
state of his home solar battery power system got me thinking again about taking
my own first steps in this direction.
I mentioned previously that I thought I might think about an emergency power
solution to get me through prolonged outages (as part of my general 72 hour
preparedness plan) as a sort of tentative first step toward one day being
self-sufficient for power, or close to it.
My plan for this has always revolved around 12V lead acid batteries, which are
cheap and simple and safe, compared to anything else. I want solutions I can
understand and repair myself from scratch or as close to it as possible. LiPo
batteries are great, but they are own practical if every battery has a tiny
computer in it, constantly and obsessively monitoring everything about the
battery to keep it working safely. That's pretty much the antithesis of what
I'm going for.
I always thought this would provide a good excuse to master the MC34063 chip,
which I bought a couple of cheaply a while ago. This is a dead simple and dirt
cheap "jellybean" DC-DC converter chip from the 80s, which could be used to turn
12V into 9V or 5V or 3.3V as needed for various applications, without the huge
wasted power and heatsink requirements that would come from doing this with a
linear regulator like a standard LM7805. The MC34063 is by no means the best
chip for this, but it is a simple, classic part, made by many different
manufacturers. You'll be able to scavenge MC34063s pretty much anywhere on
Earth after a hypothetical postapocalyptic collapse of society, so they're a
good thing to know how to use well.
However, during my thinking and reading on this, I realised that there's
something that can be built entirely with off-the-shelf parts and very little
electronics knowledge or skill whatsoever which could potentially be really
useful for this kind of application. Not that I am opposed to acquiring
electronics knowledge or skills, on the contrary, I love it, but I also get
excited about simple "everyman" hacks that are widely accessible.
The idea is to buy a cheap 12V SLA battery, a car cigarette lighter socket (the
female part), a fuse, some cable and a decent quality car USB adaptor. None of
these things should be hard to find or terribly expensive for most people, and
they're dead simple to put together into something which could be used during
blackouts to recharge phones/tablets/etc., run a RaspberryPi or similar, or do
anything you can do with a working USB port (and Ikea sell a dirt-cheap USB
charger for NiCd or NiMH AA batteries which I have heard good things about). If
you keep your battery topped up from the mains power in storage, you should be
able to get a few emergency charges of your 5V devices during a long outage, and
then you can recharge the battery from the mains afterward, ready for next time.
This seems to me like a pretty useful and practical DIY project. Of course, you
can then look into ways of recharging the battery without grid power if you
want, so it's kind of a gateway project to off-grid life.
I'm fairly excited about this, but at the same time, I wonder about the real
utility of it. Part of my 72 hour prepping threat model is to assume that all
communications utilities are down for those 72 hours. As part of this I have a
multi-band radio and a stash of spare AA batteries for it, and hopefully this
will provide me with any important news during some kind of diaster scenario.
That radio will happily run for weeks off one set of batteries with standard
usage, so relying on it for 72 hours doesn't actually require anything like this
at all. And if the internet is down and the cell towers are down, I don't
really care all that much if my phone is fully charged. Well, the GPS
functionality of a phone or tablet could be handy if the situation escalated to
involve evacuation, but for the most part, if the networks are down my phone is
basically useless to me. I have a fireplace in my house so I could burn wood to
keep warm during the winter, and I have an alcohol burning camp stove I could
use to heat food or boil water. The only thing I would really need electricity
for at all which could plausibly be provided for 72 hours by something like a
motorcycle battery is light. We have LED headlamps which take AAAs, so I guess
being able to charge AAAs would be useful, and it might sometimes be nice to
have something that could light the whole room. Maybe I should look into
rigging something up with those superbright LED stars.
It would not be much use in an emergency, except in some wild fantasy where
computer geek preppers exist in sufficiently high density to support some kind
of mesh network, but I am curious how long a Raspberry Pi using WiFi could be
run in this way. Wait, why am I talking like this is some kind of empirical
question I can't figure out for myself? Some random Stack Exchange post[2]
suggests an idle RPi3 draws 1.15W and fully-loaded draws 3.6W, so let's say
something like 2W mean. That's 48Wh per day. Let's ballpark the efficiency of
our car USB adapter at 75%, so the battery has to supply 64Wh per day. At 12V,
that's about 5.2Ah per day (if I stuffed any of this up, please email me!.
Without shopping around at all, I've found a 5Ah 12V SLA battery from a Finnish
webstore for 25 euro (about 30 USD), not much. Of course, you can buy much
larger batteries, and you shouldn't discharge SLA batteries completely if you
can help it. Let's say you were disciplined and computed for 2 hours every day,
you'd use 14 hours of power in a week, or about 60% of the capacity of that 25
Euro battery. That's not bad at all.
Of course, this doesn't take the *display* into account, which kind of shoots
this whole thing down. Hmm, people were talking about using eInk for terminal
devices a little while back in the phlogosphere, weren't they...
[1]
gopher://sdf.org:70/0/users/melton/phlog/02042018
[2]
https://raspberrypi.stackexchange.com/questions/5033/how-much-energy-does-the-raspberry-pi-consume-in-a-day
