Technically what you are describing is not a “heliostat”.
A heliostat re-directs light to a stationary and fixed
receiver. You are describing a single axis, or dual axis
if you control 2 axes, solar tracker. You are attempting
to keep your receiver “normal” to the sun.
I.e, always directly facing the sun.
This is not called a heliostat.

A heliostat redirects light to a stationary receiver.
Furthermore the geometry between the sun, mirror, and
receiver constantly change throughout the day. OK, a
special subset of heliostat geometries can redirect light
along the polar axis and generally are called “Coelostats”.

What you are describing is generally called a single axis
solar tracker. You are using a stepper motor to effect the
movement. This is laud able. I really admire your ingenuity

I ‘m a member of a BEAM robotics group. I and Wilf Rigter
have described a super simple stepper motor solar tracker
driver for stepper motors. See:
http://www.redrok.com/beamcircuits.htm#7e2
This uses a single IC and 4 transistors.
OK, double this for 2 axes.

Some have asked about the use of LEDs as light sensors.
This works very well. I use clear cased green LEDs a solar
sensors. Green LEDs can produce about 1.7 volts. This is a
lot more than wimpy a silicon sensor. 1.7 volts can
directly drive CMOS logic gates without the use of OP-Amps.
Cool huh?

All semiconductor devices can act as light sensors.
The voltage is defined by the “color” of the light.
Einstein got a Nobel Prize for defining the relationship
between the color of the light and the resultant voltage.
Called the Einstein photo electric equation.
Silicon is about .55V.
Red LEDs at about 1.25V.
Yellow LEDs at about 1.4V.
Green LEDs at about 1.7V.
Blue just under 3V.

These exotic semiconductors are not cheap. So they will
not be used in solar panels. However they work just fine
as light sensors.

Duane

I have also been inventing some heliostats that use phototransistors or cds cells to determine the direction of the sun…not unlike the floppyostat. here’s the link if you want to check it out

http://www.heliotrack.com

However the multimeter reading will be off, FWIW. Multimeters generally have a input resistance somewhere around the 50K-100K and that brings the voltage down. The only way to know how many LEDs/detectors to stack up on a CMOS gate is to test them directly. With sunlight possibly you could get away with 1 yellow — I used two because as a night owl I needed to test on a lightbulb and one wasn’t cutting it.

BTW, in case you guys want to know, so far you’ve chewed through about 2/3rds of the 1G filehigh bandwidth I offloaded the pictures to and there’s still > 1 hit a minute. You people really don’t sleep do ya :-) No worries. Next stop is photobucket.

http://72.14.203.104/search?q=cache:FkBXvbTB7FkJ:abrij.org/~bri/hw/sunflop.html+&hl=en&client=firefox-a

There is a google cache or it if you can’t load the site