Oliver Schalles 24.07.2018 16:05

Project 'GiraSol' - Oliver Schalles


Calle Calvario 15
04288 Bédar / Almeria
0034 664025031



8th. Update,  from 23. 11. 2018 @ 17:30:

The  optical endswitches...



the  End-Switches…


as recognized - the mechanical switches I  wanted to use and installed them already one year ago - they start to rust, to oxidize - so a more durable solution was in need.

mechanical endswitch


I  googled around in my pile of electronically parts  and  found - some  pair of optical  devices:

an infrared diode mounted in opposite  to a  phototransistor - so the light beam opens the  collector-emitter path.. ( so  a ‚ positive logic‘ )

But, to toggle the electronic in the Girasol -Electronic Box  - it's expecting  a  signal that goes logic 'low', '0‘, so the signal had to be inverted…


The final 'switch' is to toggle the entrance (an RS Flipflop ) - the active path - an 'open-collector' draws the input to ground, to trigger ..


I need to build 2 pair optical end switches - one pair for the elevation axis, and one pair for the rotation’ ..


first schematic:

first schematic m


The detecting electronic works on breed board:


At right the  'interrupt to the 'light beam device'

The led indicate 2 channels, and one I grabbed by the oszi..

Its a bit lame, but it works perfect for my small issues - and was 25$..  (!!)

You can see, that the logic level changes from high, ( about 4.2  volts, ) to low, about, 0.7.. 0.8   volts..

Its set that one step on the screen represents 2 volts..


Normal ' MHz oscilloscope - one channel, starts at 200$; good ones, with two or even 4 inputs - at _500,  digital storage, too, 1000$,

( talking about hi-end,  - than open end..  goes up at  6000.. (! )

So that device ' I have is a super cheap one, digital, stores as well, all basic functones available.

And I do not megahertz, just to see if that signal chances, and how !   You can see roughly the power supply, if the power is clean '

And that's basically you need for the hobby


breadboard,  one pair:

This little board is the soldered version of the breed board ...

I need 2 of them  - one for rotating east/west and the other for the elevation - horizon/zenith:
solderd version of breedboard

When running it shows the state of the optical switches

Green - no interrupt, Red - beam is blocked, (device is at end point..)

Sharp , small

connected to a pair




the  second device,but now with ic socket.

I removed  unnecessary pins and left over  the 'active pins in the socket' s


the TD62503 consists of 5 transistors - the base as open 'input, and an open collector 'output, basically - 
and all emitters are connected internally to pin 8..
Only  4 of the transistors I used : the 4 inputs -  and 4 open collectors (200 mA each) - and the common emitter:
TD62503P schematic

the second  device  with ic socket



connected  to :

connected to a pair of big sharps'

The yellow/green gives the signal; Red - positive for IR Led, black  - commune negative, ground...

So have  2 pairs of  detecting boards with the 4 sensors -the return points at east, west, horizon  & zenith


When the machine runs 'into'an (optical) switch at its  end'point, it has to stop and do reverse (by the GiraSol - Program ):
these are the detectors, to tell the program and 'it asks - is there, at that pin, connected to Arduino -  a high ( 5 volts,) or low' - logic, = 0 volts ??


The following electronic, directly after, is to store the way  (direction where to go )  - a toggle - RS FlipFlop tells', stores the state - if east  or west switch is ' triggered, it changes its output state:  on its tour to west = high', going to east = low'..

West tour ends' up  - at that detector when it runs into, 'triggers‘ the switch, and the prior high signal now goes low ... = now, go east,  - until you' hit the other end - the east switch..


The signal cooking, signal shaping' and toggling, does the hardware before the arduino..

The whole action is focused on - if one of them 4 reaches the switch, the toggle action is performed - the toggle line is connected to arduino, and the other toggle signal - as well to the Arduino. One of these signals ( high/ low) triggers the 'Service Interrupt Routine', ( at pin D2/3 )  witch calls a routine, wto look what state that toggle signal has:
all members in the program is now informed, that the 'goto-west' direction has changed - now it shall 'go east'.


for having 2 motores, all has to be double, so 2 interrupt pins, and 2 toggle signal pins..

So the need of 4 detectors…


so far,  theoretically …


Next day I found out, that the strange behaviour (sometimes a randomly’  triggering)  of one pair I recognized - at test phase with them big' ones, -

was due  a mistake I made:

I assumed that the devices were build up similar, consisting just one  IR LED and a Phototransistor - but after a check and download the data sheets -

I saw that the bigger ones were completely  ready   - no need for a resistor for the IR LED:
just to  connect to  Vcc, Ground  -  and having than, the open collector to use ( iCE  @ 40 mA ) ….


The 'small' one from sharp is the simple version, diode / photo transistor.. 
the  simple ones

- the big one has it's electronic' already built-in:

So I needed to skip, simply, just short out' the 350 ohm resistor, to sufficiently power up the device.


The resistor - limited the power, the device needed to run fully: so that strange behaviour is/was 'caused by not sufficient supply voltage..:

small and big sharp's



the cables  have  to be  well 'isolated -  not prior against electricity  - but against moisture..
to identify: the + pole is the shorter one ..

short pin pluspole

now, than: measuring cable' length, attached some jacks, designed the PCB where the two pairs of signals have to go in,

to be shaped and amplified, to go to the GiraSol-Box'


A spiralchord from an telephone gives free movement for the senselevation detectors, of course, they are moving with the mirror, so there the rotation'..
telephone spiralchord


Now, the last 3 days  ‚in’ to build a lil board that does, in fact, trigger the switch inputs of the GiraSol-Box,

Had to open the box to see if my cable to them devices having the same color-lines connected throu'..


And that thing, the attempt’ was OK, but did not work to trigger:
attempt with 4093

So I've had to change the design, and found a even better solution :


now the final’ switch/trigger/contact to the  RS FlipFlop -  will be done with optocouplers ( its  IR Diode  in series  with the visible LED) The open collector,   draws the input 'low, perfect to trigger !!!

It does it’s  work,now,  yesterday evening until this  morning  -  reliable, all states  were fast and well recocnized and super stable':
schematic  of optical trigger channel

Right, bottom, the cable jacked in'to the GiraSol.box...
new trigger with optocouplers

the  6 wire cable  to the output pins have to have a socket’ to the cable to push-in to the jack of the  GiraSol Box…:

Red is Vcc ( +5V)

white is ground,

Black - west

Yellow - east

Green - horizont

Blue - zenith

(Lakota colours to the directions)




It works like a charm...

Set & reset,  by 2 devices, 4 sensors, - connected to the GiraSol.Box:


A one year task: now ready, done ..

These sensors I found in a big copy machine and  had them all in my ‚scrap box‘


2 months ago, the optocoupler I used  here (4 in one) came from Farnell, from the price I won on here, on Element14 !!  ( Huuuurraaaayyyy !!!!)

So all came together at the right time..


These optical switch devices - now - has to be mount to the real machine..

Next steps lead to new updates…



warm greetings, despite from pretty fresh and wet weather, now, in Andalucia..






7th. Update,  from 23. 11. 2018 @ 14:30:

reorganizing the 'blog, for better radar'bility:

now the  updates  are  ordened chononogically:






oldest ...







6th. Update,  from 19. 10. 2018 @ 23:06:
a biig 'Thank You' to the Element14 community :


The resistors and the ic's.. :
resistores  from Farnell USA

It's just a small part of it, the other electronics I received 4 days later, after choosing from all  their  components from Farnel


I did not expect them any more - for, due to the high shipping prices, I e'mailed to Tariq - who finally did the order for me, that he should skip the resistors  -
but as I saw today,  he must have 'ignored' it, and so, the resistors were sent to me..

so  again - Thank  you  so much






5th Update,  from 17. 10. 2018 @ 20:08:
the cleaning, polishing action, with abrazive whitening toothpaste - and the diy - polisher..


the  material to polish  :

abrasive  toothpaste to polish
the tool to it  :


And, the result from today.. ( better as I thought )  - the 'reflectivity' - after1-hour polishing  -
I even could hear the different areas where it was dull while polishing'..


After working on, the sound legalized to other already done areas:

the better, finer, smoother, the better the reflective layer, every fanit scratch, or dullness, one can see, alike magnified on the result.

They say that silver layer one can add-on, thicker, to legalize scratches, but have no experience, practical, until now..  ( as it would behave as a 'filler.. I've to see )

After the polishing action, I sprayed, cleaned at first with tab-water, then with distilled..

of course - with my sprayer ...

The weather was nice, no wind, not too hot or cold and a clear sky, perfect working conditions.
Hope the weather will be good tomorrow, so I can do the final polish...

greetings to  you all - from Andalucia -




4th Update,  from 16. 10. 2018 @ 22:06:
the working double-spray-on-air-pistol' ...


as I promised I took a 'clip (this  afternoon)  from the  working' pistol' - here water - and a dry wooden board to show. :


The best - ok, it works like a charm, too - it takes from the two bottles the same amount of liquid (!!!) as intended:

same  amount of liquids

Next step's:

The last two polishing action - a toothpaste which whitens the teeth, abrasive, soft, and the last polish with normal toothpaste - even finer the abrasive,

- for not able to get here cerium dioxide, or titan dioxide as a fine powder, thought about to use talcum powder, but it has different, coarse grains in it.

So thinking about car polish liquid, but that is waxed or has silicone in it, and - looks nice, but as a bonding to the silver layer? Catastrophe.
Need to remove any faint of silicone than, and that would be more work, and disappointing, for it is not a real polish, but a masking, filling to the fine grooves in the lacquer, the dull, abrazed surface...

So I ended up with toothpaste, for its water-soluble and so for removable from the dish, no fillers, just real fine polishing action...

After the dish is finally polished, I need to clean with all I've in hand, fluoride, from the toothpaste, shall be removed with sulfuric acid - on the surface, after the acid, I'll remove, neutralise

with a strong base, then a water de'tensifier, and distilled water to clean all up...

I can see how the molecules work to each other !!

the last two days  - a lot of rain, here...
greetings  from Andalucia





3rd Update,  from 12. 10. 2018:
how to 'mirrorize', to make the dish 'highly' reflective -
the  need  for  a double-spray-on-air-pisol' - and the DIY build:




At first:

I have to, want wholeheartedly' say a  biiig "THANK YOU" to all here, in the community,  for the price I won - and for the help I received ...

I could choose the electronics - and all came to my hands -  and it felt like having Christmas!

It was  really  thrilling for me   

I will go on with the updates - likewise logical - the last entry, down'under' was:  - how to silverize that dish-  and found out, that I do need a finer mist to spray on. -
this garden-pump-sprayer - they made too big drops, and the handling was pretty challenging.. (the liquids from each bottle has to be sprayed on, simultaneously ! ) 
so I looked for an other, better solution since than.

My neighbour Christobal lend me his air compressor - with a dust blower pistol, from which I unscrewed the tip -  and looked around in my boxes of material' and ended up with some  connectors and rings  - from the garden hose. The connector, to attach the end of the hose - fitted ' accidentally ' perfect in the pistol: I even could screw it on'- with a ´bit of force, so - it was perfectly tight.

parts to the double-spray-on tool

now - I could use the pistol it to fit anything on it!  (a ' female' connector)
after that 'important step'  - I looked further on - the male connectors to it - found two -  which I could screw-on each other.
One of it I extended the outlet - with a PVC tubing,  some  8 centimetres - to have the outlet for the air.

My idea - then: letting end fine tubings at the outlet, so the air will take the liquid out and form the spray...
to add in 2  little tubings from a  venofix butterfly (the med-doctors use them to get a blood sample from off your vein') .
I just wanted to let them end, in the endpiece...

venofix butterfly

but how to get them in that housing?
I drilled  2 little holes on each side of the endpiece - oppositional' than I cut off the needles and pushed the little tubings into the housing - one left, one to the 'right -

and long enough to end up in the nozzel. A drop' of CA glue, tightend the drillings... :

housing of the tubing


the end-piece:  all parts in place:

the first attempt , the first endpiece

But, that 'construct' worked - for just around 5 minutes:
The air stream forced the ends of the fine tubings to vibrate -  and so the ends destroyed themselves by hitting to each other ...



The need to reconstruct  the entire 'end-tip':
A concrete airflow, a housing for the end of the fine tubings. Form metal. Glued, fixed together - like a pack of three.. :

wow - and again,  hovering the house, was looking in all boxes, cartons. And I found' a radio antenna. The last end part was fine,  but was it thin enough to fit?

reconstructing the nozzels


tube,  red shrinktube, metal tube,  the end of the flexible, a toothpic in the lumen' :

tube,  red shrinktube, metal tube,  the end of the flexible, a toothpic in the lumen'

Cutting, measuring, recutting, until all the 3 ends were in place - the tubing outlets for the liqids and the air tube. Before glueing the fine flexible tubes, the ends - into the fine metallic tubes, I pushed in the ends, in the lumen a tip of a waxed toothpick - to prevent that any glue would clog up the fine outlet.

ready  for glueing


airtubing and the 2 liquid dispenser tubings - that tiplet - to get in place:

airtubing and the 2 liquid dispenser tubings formed to get in place


here in the housing, starts the airtube:
airtube starts here..

After the ends were fine and cured I added the longer air tube - fixed altogether by shrinktube, some epoxy to the triplet and all at, in' the end of the PVC nozzle...


shrink tube forms the triplet :

shrinktube forms the triplet


a bigger shrink tube all around, to tighten all up and to stabilize. - the shorter tip is  the air nozzle (red) :

glued in and  sealed  with sh'. tube around


all in place, ready to connect:

all in place, ready to connect

That construct works like a charm. It draws about 1 ml every second to nebulize... 

Now,  how to attach the little tubings to the liquid bottles, and a holder for the bottles -  to that pistol ?
browsing the house, an empty 5-litre demin water canister would do it ':

cutting away the half of it from top to down, cutting away the unused parts -

at top of that construct a big hole to fit on to that pistol, to the garden hose connector... :

into the open syringes I can add some filter material, and they guarantee that the tubings will be well 'supplied, always submerged in the liquid..
the bottleholder, 2 x 0.5 liters:

the bottleholder


the center drilled caps with the superflexible liquid-tubings:


after all testing all has to be cleaned, all, the working marks...  yes...

next step - waiting on good weather, to clean the whole dish thoroughly, to give last polish on it, and than, the setup of new  silverize-fluids -

the  activator - ( tin chloride );

the  silversolution itself  (silver ions  bound in 'aminoacid-complex')

and the  recductor - (glucose and a vitamin C, organic  acid )

further above, in the previous update, I posted the more complex receipt to it.
the tin chloride, the activator,  -  I will spray on with the pump-sprayer. 
and for the reductor and the  silver solution, I will use that new -


so long - i ll keep you updated !!
with nice  regards and  greetings  from Andalucia -







2nd Update: from 23.09.2018
the history  about  to choose  the dish, and  - how to 'mirrorize', making  it 'highly' reflective:

All them 'parabols' I had im my hands were coated, of course - the one I used here, shown in the video, has a core from aluminium with is massive 'coated'  glass fibres X epoxy (structures on the backside)  and a  fine layer of it on the front..

I wet-sanded the front, with 150, 300, 500, 800, 1000  up to 1.500 - the parabol' was  laid up ward, and I filled the parabolic gap with water and some drops  of a  dishwasher, detergent'

My grinding tool was a 30 cm round tap from a small rubbish bin...and I just followed softly (!!) the form, curve -  not to change the geometry...

than the question:  which way to use - to make that parabolic area highly reflective?


Aluminiumfoil ? , space blankets ( mylarfilm, golden and silver reflective )? adhesive reflective foil? silverize by chemics?  wow.

a loot of ways,  one as bad and unpractical, as cutting the foil in stripes, glueing them, step by step, to cover the entire area, (glueing?  - with what? - a glue that contains thinner in order to evaporate to cure'  - how will that solvent go through mylar, a tight foil ?  - no way. , so 2  components epoxy?
Idea good' (no thinner need to evaporate) - but:
Wow, what a procedure  - to get the glue on, in a fine coat? You see every irregularity in/of the geometry of reflexion: even the faint layer of epoxy ruined the surface... 
And, additionally  - after a  year, the entire work starts to get dull on the cut edges, and the mylar film 'brittled' loose, and leaving the metallization randomly on the surface: very, very bad. - Frustrating-No-Way. ...

Than, finally, I found a 'patent' to silverize 'ANY surface'  as they claim - it made sense to me - and I started to achieve all the stuff I needed for...

Made the liquids, and tried'out  - on a  glass plate, to see if the chemismus worked - and it did.

The liquids need to be adjusted in concentration, and I found that the pump spray devices (garden center) do spray, work - but the produced drops are too big, so I need finally an air-compressor  - for a finer mist ...

The silverizing solution and the reductor liquid has to be sprayed on, simultaneously and while meeting on the surface, it forms pretty immediately a fine reflective silver layer -  from 10 µm to ...150 , 200 µm, depending on low long you treat the area.

On the  glass plate, I found areas, where the  silver layer was tightly  bound to the  surface - and  areas,  where  I could  just wipe the  layer off, with my  fingertip, but it was not intended  to silverize, its intention of the probe was to see if the liquid' work as intended...)

It is highly depending on how clean you (can) work. Every fingertip, every particle, or a faint residue' of oil, grease, fat, will hinder from bonding..

The liquids have to be made with demineralized'- or even distilled water, all rinse-procedures have to be done as well of course-  with dest/demin water ...


the 'receipt: liquids to silverize...
the 'receipt: liquids  to  silverize


Activator, reductor  & silver in solution (light sensitive ! -  so I wrapped the bottles in alufoil...)
reductor  & silver in solution...


the  3  pumpspraybottles -  at the  location of the  test  - outside in the patio:

  (and  it silverize nearly  all, but, a bit nitric acid, diluted -  will swap it all clean again ...   )

the location of the  test - in the patio! the three pumpspray bo


overall silver ! - here  to remark: at the  areas  where the  activator  could  reach - there, the  layer  is  very,

very  tightly  bound -  even to ceramic... !

overall silver !


the outcome from the first test, a  spray on, some 4 seconds.  So the layer is ultra thin and the object has to be highly clean before...

the  3 liquids:
1.: the activator contains stannum, tin ions (tin chloride) that gives any surface the possibility that the silver will have 'starting points -

from where the micro crystals do grow from - to form the metallic, conductive silver layer.


2.: the 'reductor' is based on organic acids, glucose (kind of sugar), and vitamin C


3.: the silver solution -  the main ingredient: Silver as 'nitrate (water soluble) and an organic complexer' (Glycine, Amino'aethanol) that 'carries' the silver-ions..
When the reductor and the silver solution meets at the surface - the silver (iones) will be reduced to metallic silver - the reflective layer... 

so far the metallizing actions - when I do spray it on the parabolic  - I let you know from, of course ..

when once - the reflective layer is applied -

I have to clear it out, even to polish at certain areas - and then there is the need for a final UV protective, clear, fine lacquer coat,

witch the silver protects from atmospheric influences...


my  facebook

account - if someone wants to contact me there, too,  you are welcome!




1st Update: from 19.09.2018
the 'new construction to the elevation axis -



the force  was so strong

that  the old rod nut  was

pushed out of its harness...

wow.  - I was looking for a new possibility to order the entries -  found until now ' nothing -  so I will append the material I do have.
Today  I  completely  was  able  to renew the elevation 'system:
the motor, the connection from the motor housing, case to the bear 'ring'housing was wobbling, so I used just a piece

of a wider metallic tube and resined (epoxy) the gap in between the housing of the motor and the bear'ring housing... perfect tight and rigid, now. ( some days ago ) ..
The rod nut, in its housing ' rod-nut-holder' -

a long 'nut, hexagonal, glued in - as well with epoxy. To make it  'fit in - I widened the existing channel, and

dremel'ed in, the hex profile,  so

the  long nut finally fitted in..

As  well some  days  ago

and  the


' work: the strengthening to the  sides, which carry, will hold the rod nut'-holder...  (the  nearly  triangular pieces of metal, with the

bore hole

for the  axis) - glued to the


of the down-under-part of that parabol.. :

Here to see the motor, the rod and the housing of the hex-long-screw-nut', and the triangular strengtheners':

the  motor, the  rod  and  the hexnut - housing

the hex-nut, glued in the holder (epoxy)


the triangular strengthener'
glued-in (epoxy), strengthening (2 mm) metal plate..


does all fit to each other?
does all matches together? ;)


close'up of the hex-housing'
a  closeup of the rod  witch  'enters in the axis  of the hex-long-nut housing'


aside: the nut's, M6, Inox
the strengthening triangels. (nut : inox )


the motor,  the black tube to the  housing,  and the silvery tube-housing of the  bear'ring:
the motor to it: the  black tube is the stabilizing, the silver' is the house' of the bear'ring

a  shot from below, upwards: motor, rod, hex'housing, attached  to the bottom of the paraboldish;
      above the motor: the elevation axis..
about 45 degrees to zenith, now :)  a shot  from below..

.. about 55 degrees
clockwise turning = looking ' Zenith; counterclockwise turning the  rod

last year, what's  today's  left over - from the polishing  action ( without the  reflective  silver layer  !)

the laser beam reflects at first spot, to the second, to get reflected to the third. :
(5mW 405nm):

that was  amazing  for me too -  a  ok, polished  (a year ago,) surface to be ready to get metalllized, coated  with a reflective silver layer.  The  laser  beam is reflected, at first,  to the  second spot  which lead the beam to the  third... Laserpointer


horizont' position - at  evenings and in the mornings.. (we  will see  )

horizontal' position - in the  mornings ( ?) and in the  evenings ?  - we  will see... :)

the  coloum cacteea in background are  Trichocerus - terchekki, -peruvianus & -panachoi..

I was totally overwhelmed - that more than 13.8 K visitors had sawn my project.
  and - I do 'Thank You' for your time - to see, and look and follow, or even comment..

A good day,  a good night - wherever you are - hope you are ok,  and  well -

greetings  from Andalucia, con mucho sol -
- Oliver





My first entry from  24.07.2018 16:05

I am completely new here (joined today) ...

... and  have no idea about the  'what's about',  but I do work on a  'heart' project  - my  'GiraSol' Project, which I wanted to share  with  you :
its a 'device, consisting of a  reflective parabolic  dish, that collects  sunrays, energy  to a  focus  where a  converter' (copper pipes in where  a fluid  gets  heated  up and  is  lead to an isolated boiler, where the  heart is  given into it, to the usable water - for  a hot shower  or to use to wash up the  dishes  in the kitchen - it even could  be used  to heat up a  floor in the  house...
the  sunfollwer electronics, Program to calculate  is  made  for an Arduino Nano, it has  8  analogue 'In'put pins :

5  for them photodiodes, 3  for the temperature from the boiler

2  big motor driver ic's IBT ( h.bridges  for DC' motor's ), to drive by  PWM the rotation motor ( east-west )  and the  elevation motor ( zenith-horizont )

the  sensors are made of 2 pairs of photodiodes, and a centre PD, all in all, 5.
4 End-switches, to detect the max'mechanical limits  - after 2 Schmit-Trigger RS FlipFlop - polled' to  2 interrupt pins..  ( on D2, D3 @ Nano )

The  thermor  has got its own ArduinoUno - 
for the 3  NTC - and the mechanical overtemperature switch

the temperature - coming from the  Parabolic  Dish, to load the thermor,

the  water'temperature  'absolute, in the thermor, and -

the temperature at the out'let  - back to the focus,  to get reheated there again...


OpAmp followers  and  one comparator, to detect overtemperature,

a 0.9' Oled Display to show the current temperatures -
a pushbutton to show up the data on b.press..

I have  a load of photos already posted on Facebook on its  'project page - here: the link to it:
shall i do upload them photos / clips  here too??  and if yes - where to ?  - I definitively need help not to get lost here ...

Greetings  from sunny Bédar, Andalucia, South Spain..





first  steps - 2  photodiodes, arduino nano, the  sketch - listed  below -
the motor  driver, the ac windowlever motor, connected to the rotation axis:
(a rear-axis  from a opel astra, without all them mounted  stuff on it.. ) :


Overview:  in front the  Parabol, in the back, the thermor in which the energy will be stored in'..

a test to  pan, tilt the  system, to see how it responds to different voltages:
dc motors do reverse the direction of rotation when we do reverse the connections of the dc voltage.
the speed is pretty  high  with  current, steady 12 V DC - but in work modus',  the  speed is  really, really slow - realized via the  PWM... '
PWM: tiny spikes of  12  volts -> very slow motion;  5 ... 10%
longer, wider  spikes: half impulse <>  half delay' - forces to spin faster: 50 %
and real big ones,  long impulses witch nearly flowing to each other:  the motor spins really fast ...  90%:
These impulse and pause times were repeated in nearly 1:1.000, 1:500 seconds interval.. (Arduino et al. ) depends on which pin is  assigned.

the  GiraSol-box - the  Arduino Nano as the core - and two  mighty dc motordriver for the  asxis's: