Re: Q. re: ratings for short circuit and open circuit

ehsjr <e.h.s.j.r.removethespampunctuation@xxxxxxxxxxxxxxxx> wrote in
Sa4jk.26$vu.10@trndny05:">news:Sa4jk.26$vu.10@trndny05:

Kris Krieger wrote:
ehsjr <e.h.s.j.r.removethespampunctuation@xxxxxxxxxxxxxxxx> wrote in
t9Iik.25$vu.21@trndny05:">news:t9Iik.25$vu.21@trndny05:


Kris Krieger wrote:

[much snippage]

[...] I might be able to get away with 400mA,
but I'm not sure.

Ok. You can convert the voltage and current numbers to a single
power number (watts) by multiplying volts times amps. Power in
can never be less than power out - you can test your initial
design against that watt figure without ever buying a part.


DOH! OK, so that means, to get that 5V at 500mA Out, if I have 1V
in, it has to be at 2.5A (2500mA)?

Exactly (mathematically). In fact, you need a little more input
power than output power, because there is no such thing as a
perfect lossless circuit. The circuit that converts the 1
volt input to the 5 volt outlet will waste some of the power
as heat, so you might need 2.7 watts input to get 2.5 watts
output.

Ah-ha...

[snip]

OK, so, W (for the given circuit) is 2.5 and W=IV, so I=W/V Ok
then, so, to get those 5V going out at 500mA, using this specific
voltage-boosting circuit, what I'd need (to get that output) is at
the very least one of the following combinations:

0.5V at 5A
or
1V at 2.5A
or
2V at 1.25A
or
4V at 625mA
or
some other combination where V is between 0.8 and 4.5, and a current
I such that the resulting W (power) will be at least equal to the
output power of 2.5W...

Yes. :-)

WOO-HOO!!!

OK, more things are now making more sense. I'll have to go back to some of
the simpler circuits I'd downloaded and printed out, and take a re-look
with that in mind.


[...]



I'd read about the pulsing, which sounded like a great idea, but I
haven't yet figured it out.

I know I want the LEDs to run for 8 hours. So then, what I

Ok. You can convert, again, to a useful figure.
First, you want 6 3.4V LEDs to run at 25 mA - that's
.085 watts (3.4*.025) per led and .51 watts for 6.
Next, you want them to run for 8 hours - that's .51*8
or 4.08 watt-hours. That means your solar cells will
need to provide at *least* 4.08 watt-hours every day,
even if all the electronnics was perfectly lossless. In
practice, you will need a lot more than the minimum.

Oh! OK, the cells have to produce, let's say 4.5 Watts (it's a "mor
round" number), becasue that has to go into the batteries, in turn because
that's at least how much the batteries have to be ablet o provide to the
LEDs ... and the electronics designer uses the subcomponents in the circuit
to determine the Volt and Amp balance/numbers/proportion needed to be suer
that the conditions (4.5W) are met...


Now the hard work starts. You need to determine the average
power per day you can expect from your solar cells. Then
you need to consider how many days you want your lantern
to function when the batteries cannot be re-charged by the
sun due to rain/snow/whatever. You also need to figure
that seasonally - more power will be generated in the
summer than in the winter in the northern hemisphere.

To be honest, that's why I was looking at the battery-charging IC - I
figured that, if the maximum Volts and Amps of a given set of solar cells
can charge the batteries in 4 hours (IOW, under bright, clear, sunny
conditions), tehn that same system ought to allow the batteries to charge
OK if the unit receives, for example, only 5 or 6 hours of sunlight, or, if
the sunlight is degraded due to suboptimal weather. THat's about as far as
I could go with it. So the Bbattery-charging IC was supposed toprotect the
battery from over-charging if the unit received more than 4 hours of
sunlight.

I don't know if I'm describing that in a way that makes sense. But the
math of it is too daunting for me (and I don't have any test equipment, so
I'd have to buy some to test a cell under all weather conditions - so I'm
admittedly "winging it" with the charging time).


Your worst case figure (ie lowest average power over
x days) will have to be at least 20% higher than the
4.08 watt hours per day, because you need to charge
batteries at 120 percent of the amount of energy used,
due to losses while charging. And, if there is a voltage
converter circuit used as in the ap note, there is loss
in that, too.

Oh!, I didn't know about that 120% - I'd read several battery information
pages, but I can't say whether they didn't mention that, or whether I just
forgot due to Information Overload. Well, it's now in my Notebook, at
least.

((This si something of a mini-education in the difficulties associated with
alternative energy...))


[...]
- the booster puts out 5V, so I'm still trying to figure out how I'd
need to tweak the example charging circuit to figure out what
components I'd have to change for it to work properly with the lower
voltage. This has been eluding me (becasue I still just don't know
enough).

Well, if you look at the "garden" lights (I know you have) available,
it's been eluding everyone else, too. :-(

The thing that possibly saves you is that you can charge more $ for
your product than those cheap garden lights. The thing the limits
you is the amount of electrical power you can get from solar cells.


Right on both counts ;)

I am just convinced that eventually, I'll work somethign out,

From what I have seen of your interest and effort, I agree.
You're not like some who ask a question and disappear - you
dig your teeth in and work hard on this. Most commendable!

Well, people who know me woudl probably say "maddening" ;) Seriously,
tho', I typically "leap before looking" and have to guard against
frustrating myself (since frustration makes me cranky). But I am
particularly (or perhaps peculiarly ;) ) motivated to put my best effort
into doing this, becasue of the broader implications/ideas that I see
associated with it.


and I think
it could expand both the field of glass-work (I also have soem ideas
for using what's called "warm glass", where you melt different
colors, shapes, etc. of glass together and then use heat again to
shape that) and the whole idea of solar lighting.

Then there is a new invention:
http://news.cnet.com/8301-11128_3-9988923-54.html?hhTest=1
http://web.mit.edu/newsoffice/2008/solarcells-0710.html
whereby windows can be made to fiunction as both windows AND solar
collectors. SO I can envision combining this invention with methods
used to make glas shurricane/impast reistant, to create self-powering
'light tubes" and other forms fo public art that can also function as
self- powering night-time lighting.

Nice links & I like your ideas.

Thanks :0 I actually sent the "light column" idea to the fellow listed as
the lead researcher of the solar-collecting window project - I'm sure I'll
never get around to any sort of patent, so might as well at least pass
along the idea.



Of course, what I need to do is, as always!, rein in my over-active
(and, yeah, sometimes grandiose) imagination, and succeed with the
more modest practicalities <LOL!>

Groan - I know exactly what you mean.

Heh, yeah, the imagination can certainly run way out ahead; it can be
difficult to self-check and be sure it remains rooted in reality. OTOH,
progress is made by pushing the envelope, so it's also good to not get
*too* rooted ;)

As to the practical - I have an alternative for you to
consider, but it may be based on an incorrect assumption
on my part. My assumption is that you want your lanterns
to illuminate something other than themselves.

Yup.

My idea
is to use LEDs to draw attention to the lanterns, but
not use them to illuminate things. The electronic difference
is 2 mA LEDs at 2 volts vs 25 mA leds at 3.4 volts. And that
idea designed itself. By that I mean it selected a particular
solar cell (6"x6", 10V at .15 mA) that can charge 4 NiCd
1500 mA hour batteries directly through a diode. No expensive
or complicated electronics to charge the batteries, and a very
easy energy budget - .004 watts per LED vs .085 per LED.

The penalty is a change to your basic idea, if my assumption
is right. It's bright, pinpoints of red light versus much
broader white light as I think you have in mind.

Well, I have certainly been thinking, over the past 3 or 4 days, that the
"six LED" idea is probably overkill. I am stuck on the white, or at least
warm white, tho', not only for lighting reasons, but becasue it's the most
popular color, also it won't clash when I move from clear textured glass,
to colored glass.

The lower-power red LED ( "capacitor rather than batteries" OR "capacitor
in place of batteries" OR "capacitor instead of batteries")ecially in terms
of the energy-budget, is reasonable, but colors are dicey in terms of
selling.


The point here is not to push my idea, but to indicate that
there are practical alternatives the will work, even if it
turns out that your energy budget won't support what you want.
Those alternatives "penalize" what you have in mind in order
to reduce the expenditure of energy so that your energy budget
can support a different version of what you have in mind.

All that is true. I've been thniking about things like (1) LED
light/viewing angle, and (2) if the units are to be primarily decorative,
maybe it'd be better to do something separate, or at least, larger, for
brighter lighting. Larger, of course, means a more robust structure.


Take pulsing as an example. Suppose you turn your 6 LEDs on
and off 100 times each second, such that they are on for
a total time of 1/2 second and off a total time of 1/2 second,
every second. In 8 hours, they would be on for 4 hours, and off
for 4 hours, but they would appear to be on for the entire
8 hours. They produce 1/2 the light when pulsed that way,
at 1/2 the energy cost.

I think that th epulsing is definitely a great idea; I read a bit tat
mentioned even faster blinking. I just haven't yet read enough about it to
have any idea how to do it. I know there is an IC that flashes (a "555"?),
but I'd glossed over it becasue, at teh time, I didn't think "flashing so
rapidly that it look like a constant-on", Iwas thinking slower, noticeable
flashing. Live and learn, tho!

Your point is a good one - 1/2 the energy cost owuld be excellent. I just
have to go back and go over the references about flashing.

Re: light "boosting", I took apart an LED night-light, and there was a
curved plastic cone in it, painted with a highly reflective paint, that
focused the light, making more of it useable. So I got the idea of getting
an inexpensive mirror and cutting it so as to make a "mirror pyramid" - so,
rather than the light just pointing downwards, the mirrors woudl refelct it
back up an dangle it (the light) out the sides - that oughtto make it
appear brighter than the LEDs actually are, since mroe of the light will be
"useable" so to speak.


Of course, whether these or any other modification to the
original design are needed depends upon how much energy
you can get from the solar cells.


I'm convnced that I actually have to start from the beginning, firstly
becuase I was stuck on a track that I don't think was goign anywhere (woudl
require more power input than would be practical/affordable), and
secondly, because the more i learna nd understand, teh more I realize taht
I can't simply "stick stuff together", that integrating all of these things
requires a lot more planning (and, sadly for me, math ;) )



But that's what's behind this - it's more than just wanting to make
"a path/garden light", and sell a few. What I *hope* to do, beyond
that, is make something that will eventually inspire more capable (or
at least better-funded <G!>) people to take some of those grander
ideas, and bring them to life.


Well, this is where I see your idea having promise. Those
garden lights are feeble, and leave you hungry for something
better. You're not going to make a garden light functional
replacement. The current garden lights are supposed to
provide "accent lighting" (e.g. dimly illuminate a path)
and not look terribly ugly while doing it. Well, they succeed
in being dim, and they also "succeed" in looking at least a
little bit ugly.

<LOL!> How true. A few recent designs are interesting, but most are
underwhelming.

Your idea provides something that you want
to look at. Day or night, it will be beautiful.

THat's what I'm hoping. First one is a rectangular shape, but the design
uses clear textured glass similarly to how pen'n'ink is used, where
different combinations of lines, or dots, or otehr shapes, fill in the
design outline. I'm starting with simple shapes - one design is 5 arching
grass blades, the other is three vaguely daisy-like flowers (the market, ya
know?), panels are 10" tall and 8" wide (anything smaller would get dicey
because the leading would cover a higher proportion of the glass).

What I was originally thinking was actually making the top (with all the
lighting components) removeable so that it could also be alernately used as
a candle-holder (not as tricky as it sounds, actually).

Lighting it
at night is key to letting people see it. So, overall if it
is technically and economically feasible, I think it's a
winner.

I'm happy I'm not the only person who thinks so <L!>

I do think the technical aspect can work, but I do have to go back and re-
think the lighting, at least in terms of what type of unit design would be
best suited to what level of lighting.

I really like the idea of a "solar table lamp", for example. But that's a
different beast from a small decorative lantern.

I'm also trying to consider safety. The prototype has the textured side of
the glass inside, to keep dirt from accumulating in it, and I'm looking
into a type of clear film that's applied to house windows to make them
highly shatter-resistent. Also, the prototype will have the solar cell in
the top, so I got some aquarium sealant, and am also decifing whether to go
with encapsulated cells, or "naked" cells aht will be attached to the
underside of a clear glass panel.

The planning is certainly giving me more appreciation for people who do
this sort of thing for a living!

[snipped]


OK, I'll get off my sopabox now and go back to my notebooks <LOL!>


Damn! I was enjoying the soapbox! :-)

Ed


Which means I stepped down at a good time - *before* it got boring <LOL!>

Oh yeah, I've been googling things like "capacitors instead of batteries"
and variations fo that, and there is actually some info. SO I'm going to
look into it and see what I can see. It'd be *very* cool if I coudl skip
the batteries. Not at all sure I can figure it out, but it would
definitely be cool ;)

Back to work now!

- Kris
.

Relevant Pages

  • Re: The Full Extent of the Disaster
    ... Kinetic batteries are in use at almost all hydroelectric dams, ... The flywheel provides power during period between the loss of utility ... The pumps use electricity one way, ... They provide energy for a very short period of time, while other systems come back online. ...
    (rec.gambling.poker)
  • Re: Researching maximum life span of the power supply of a small satellite.
    ... electronics needs to generate for over 30 seconds 15 watts of power ... engineering reasons I want to minimize the use of solar cells. ... and I can't see chemical batteries lasting that long in space ... inches) will generate about 15 watts of energy. ...
    (sci.electronics.misc)
  • Re: Computer life
    ... Assume on usage power is about 175 Watts, off time is weekends + 12 hours per day. ... I was very suspicious, so out came the wattmeter, and I measured the standby power of a 42 inch LCD TV, and a Bluray player. ... Even though the energy savings are somewhat less than the environmentalists claim, once you take into account the energy costs of the actual manufacture, it totally suplants even the claimed energy savings. ...
    (microsoft.public.windowsxp.hardware)
  • Re: space travel can change the world
    ... Dash means that it's kilowatts *times* time. ... if by that you mean power times time. ... of energy and power. ... life for a particular batch of batteries. ...
    (sci.space.policy)
  • Re: Watt meter
    ... >> is watts per hour). ... >Watts per hour would be a rate of change of power. ... Watt-hours are what accumulate on the energy ... >meter, that you pay for. ...
    (sci.electronics.basics)