To: beam@sgiblab.sgi.com
From: Wilf Rigter rigter@cafe.net
Date: Sun, 27 Feb 2000 16:56:25 -0500
Subject: Re: Solar Cell tests
> Very clear and simple analysis!
> =
> regards
> =
> wilf
> =
> CyberBug wrote:
> =
> > Hi there,
> >
> > I saw your tests and I was curious about your results, so I did some
> > calculations and this are the results (or explanations of your results)=
:
> >
> > 1. When charging a capacitor the solarcell works like a baterie and res=
istor
> > in series, so to make the calculations we need to now the solarcell's
> > resistance, and that would be:
> >
> > A: 3.5v @ 6 mA
> >
> > Ra =3D 3.5 v / 6 mA
> > Ra =3D 583 ohms aprox.
> >
> > B: 2.7v @ 16 mA
> >
> > Rb =3D 2.7 v / 16 mA
> > Rb =3D 169 ohms aprox.
> >
> > 2. Now we test for parallel configuration. As you can see in the schema=
tic,
> > you have a net formed by V1, V2, V3 and V4.
> > According to Kirkchof's law the algebraic sum of all the voltages in th=
e net
> > must equal 0. So:
> >
> > 0 =3D V1 + V2 + V3 + V4
> >
> > We know V3 and V4 (the solarcell's voltage), but since they are not equ=
al
> > they have to compensate through the resistors, so we now have to calcul=
ate
> > the real output voltage. First we have to indicate de polarities of eac=
h of
> > the voltages, in a CCW fashion you get the polarities noted (based on t=
he
> > known voltages), then we can solve for the know voltages:
> >
> > V2 + V3 =3D V1 + V4
> > V2 + V3 =3D 3.5v + (-2.7v)
> > V2 + V3 =3D 0.8v
> >
> > Now we know that resistors Ra + Rb hold a voltage of 0.8v. The real out=
put
> > voltage is taken between the Ra and Rb therefore resulting a voltage di=
vider
> > between the two. Keeping this in mind we know calculate the voltage pre=
sente
> > at Rb:
> >
> > Ra =3D 583 ohms
> > Rb =3D 169 ohms
> > Ra + Rb =3D 752 ohms
> >
> > V3 =3D 0.8v x (Rb / (Ra + Rb))
> > V3 =3D 0.8v x (169 / 752)
> > V3 =3D 0.8v x 0.211
> > V3 =3D 0.179v
> >
> > Now for V2:
> >
> > V2 + V3 =3D 0.8v
> > V2 =3D 0.8v - V3
> > V2 =3D 0.620v
> >
> > Now V3 and V4 are in series and this is the real output voltage so:
> >
> > Vo =3D V3 + V4
> > Vo =3D 0.179v + 2.7v
> > Vo =3D 2.879v
> >
> > The real output voltage represents the voltage that you would have if i=
t was
> > only one solarcell. The output current I is the sum of both solarcell's
> > current:
> >
> > I =3D 6 mA + 16 mA
> > I =3D 22 mA
> >
> > The resulting solarcell is 2.8v @ 22 mA (hence faster charging, more on=
this
> > later)
> >
> > 3. Now we test series configuration. Again, as you can see in the schem=
atic,
> > this is much simpler. We now have two voltages V1 and V2, they are in s=
eries
> > and the resulting output voltage Vo is:
> >
> > Vo =3D V1 + V2
> > Vo =3D 3.5v + 2.7v
> > Vo =3D 6.2v
> >
> > But now, we have the same problem we had with the voltages in parallel =
but
> > with the current. We have two different currents, but not to worry, we
> > already know the resistance of each solarcell, so:
> >
> > Rt =3D Ra + Rb
> > Rt =3D 583 + 169
> > Rt =3D 752
> >
> > Applying Ohm's Law:
> >
> > I =3D Vo/Rt
> > I =3D 6.2v / 752
> > I =3D 8.2 mA
> >
> > The resulting solarcell is 6.2v @ 8.2 mA (slower than A + B in paraller=
l but
> > faster than A alone).
> >
> > So now we have:
> >
> > A: 3.5v @ 6 mA (11.5 sec)
> > A + B series: 6.2 @ 8.2 mA (6 sec)
> > B: 2.7v @ 16 mA (3.5 sec)
> > A + B parallel: 2.8v @ 22 mA (2.5 sec)
> >
> > What can you guess from above?
> > Right, the larger current charges faster, even though voltage is smalle=
r.
> >
> > But then again, all of the above results are theoretical, in practice t=
hey
> > vary. You were suposed to have 2.8v total, but in practice you had 2.4v=
,
> > although it did charged faster because of the current enhancement.
> >
> > Now for the charging times, if you could see voltage vs. time in a grap=
h you
> > would see that a charging curve, as you get closer to the maximum volta=
ge
> > the voltage variations are smaller and much more longer in time. You st=
art
> > charging very fast but it keeps getting slower until you reach the top
> > voltage. If the charging time was linear you would see the same time go=
ing
> > from 1.1v to 1.2v than going from 1.4v to 1.5v.
> >
> > Hope this was useful, if you or anyone has any comments please let me k=
now.
> > See ya.
> >
> > ----- Original Message -----
> > From: FCO Enr.
> > To:
> > Sent: Thursday, February 24, 2000 3:35 PM
> > Subject: Solar Cell tests
> >
> > > greetings,
> > >
> > > I did some tests with some various solar cell configurations
> > > and here is some of the data I gathered as well as some questions
> > > for those of you who can explain this phenomena.
> > >
> > > First off, I tried various combinations to find the most efficient
> > > low light (my ambient light level) setup. The winner is surprising
> > considering
> > > even my Solarbotics catalogue seems to point in the wrong direction.
> > >
> > > A - SC2422 (3.5v @ 6ma)
> > > B - SC2433 ( 2.7v @ 16ma)
> > > ( I also experimented with the SCPDs but I didn't write down and time
> > > the results so I won't include them).
> > >
> > > The goal was to see how fast the circuit could gain .01v in my ambien=
t
> > > light. I did my best to avoid any movement of cells.
> > >
> > > A 11.5 seconds
> > > A + B in series 6 seconds
> > > A + B in parallel 2.5 seconds
> > > B 3.5 seconds
> > >
> > > Now this is pretty odd considering it seems like wiring these
> > > cells in parallel improved their low light capability (not by much
> > > considering the space it takes up).
> > >
> > > WEIRD: When the voltage in the circuit reached 2.2 something volts
> > > I noticed it was takin MUCH longer.. from 2.23v - 2.24 it took
> > > 1 minute 7 seconds!!! I looked at my meter again and it hasn't
> > > budged from the 2.24v mark in over 2 mins!!
> > >
> > > I also noticed the times I stated above were between 1.4 and
> > > ` 1.58v. It seems like the charging between 1.1v and 1.2 is almost tw=
o
> > > times faster! Why does it vary depending on circuit voltage and why
> > > did the voltage sort of plateau at 2.24v?
> > >
> > > Now the light level hasn't changed AT all and I tried again when it g=
ot
> > > to 2+ volts the cycling was a LOT slower to get the same 0.01v
> > > WHY? So this also means if I get the lowest rated trigger it would
> > > improve low light capability DRASTICALLY!
> > >
> > > I would appreciate any feedback on this.
> > >
> > > Andy
> > > fco@total.net
> > >
> > > hrdware's digital sanctuary - mindless oddities
> > > http://www.angelfire.com/pq/hrdware
> > >
> > > The Factory - TF Map Making Tutorial
> > > http://www.planetfortress.com/factory
> > >
> =
> =
content-transfer-encoding: quoted-printable
>
Wilf!
You're alive! I swear=
if =
your print the BEAM Heads 101 article backwards, it spells out: "
=