Thank You, Michael! I know that, but to be sure thing would be working as I
expect, I have to do some
Raytracing before I spend money to make actual lens.

​As You can see, I made prototype from material at hand so I know thing
would work, in spite of
Expert opiniom stating this is IMPOSSIBLE.
I used sheet lens  with very fine lens on it, but here I do not need
optical quality lens, just light
concentrator. As I did not knew how they made this sheet lens, I studied it
and now I know how
to make flat one, but as You see I have varped it to be on inner face of
Dome.
I guess I have solution for that in form of simple Transformation Matrix as
I can treat Lens ridges
as 3D letters which are also originally flat,but can be put on the ball or
cylinder.
As lens consist of outer dome surface and other lens I forgot which
radiusit has to have, I would
have to experiment. It would be useful if there is some raytracing
algorithm to use it on finished
3D object, but perhaps no one tought about such things for SCAD.
If such capability would be added to test objects, it would outgrow its
simple use of 3D objects
creation.
It would be usefull also to be able to create 3D surface using Algorythms,
and it seems
to me  this actually is possible, just I do not have time to study how it
is done, probably using some
transformations matrices.
For now, I shall use what is there, even if some practical things are
missing, like exporting pieces
of larger objects as separate, best in separate .stl files, so large 3D
objects can be 3D printed on
smaller printers.
Good thing is that once I make program to draw some 3 D object using
parameters, then thing
can be scalled up or downjust by changing those basic parameters by hand
and running program
again.
This Dome would be 50m in diameter, 100 times size of this prototype.
As such large Object would take almost infinite time to be 3D printed, I
have to split it in  adquate
number of hexagonal parts, that could be then  manufactured by Injection
moulding,
It would be ideal if that can be one universal piece, but I guess I would
have to have  at least 128
different pieces, but maybe 64 would be enough if I do not mind  some
optical disorders in focus.
To be able to recognize each, I would have to put numbers on each, so they
can be assembled in
proper order.
I would really appreciate advices, and any giving them would be proud as
thea would help to save
the World by producing cheap and clean electricity from Sun and Wind, and
producing condensed
water from air as byproduct. With enough water and with Land protected from
strong wind and rain,
snow and frost and hail, much more food would be produced, so hunger would
be thing of past in
all countries with too much sunshine and not enough water.
Thanks for Your help in advance!
Marijan Pollak

On Tue, Feb 28, 2017 at 12:41 AM, MichaelAtOz [via OpenSCAD] <
ml-node+s1091067n20625h81@n5.nabble.com> wrote:

--
Regards from Croatia, the Homeland of one of greatest inventors from 19th
and 20th Centuries, Engineer Nikola Tesla!

Photo-0048.jpg (105K) http://forum.openscad.org/attachment/20636/0/Photo-0048.jpg

--
View this message in context: http://forum.openscad.org/Making-transparent-Object-tp20620p20636.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

Your arrangement can work, provided the focal distance of the frensnels is
tuned to the radius of the dome and the dome is a half sphere. The problem
with Frensnels is that they are only effective up to  some limited angle
https://www.wsv.de/fvt/fachinformationen/lf_technik/optikinfo/lichtbuendelung/bild_09.png
. So you will harvest only fraction of the available energy. Another nasty
problem that has to be solved for fabrication is the decomposition of the
dome into polygons. You'll have to use specific (laser) cutting for this.

To answer your question I don't think you will be able to produce fresnels
with a common FDA 3D-Printer - mainly because of limited resolution and
unavoidable optical inhomogenities. But it is cheap to buy such  foils made
in China
http://www.packagingnews.co.uk/news/materials/rigid-plastics/chinas-fresnel-lens-foils-to-launch-in-europe-21-12-2016
.  If I had to do it I'd rather use an engraving machine to produce a stamp
tool for foils or acrylic plates.
Given you have access to such a machine or even build one, it might be a
better idea to fab the whole dome as a single fresnel lens respectively to
fab each tile of a dome skeleton specifically as a fraction of the large
fresnel. In this case the dome doesn't have to be a half sphere.

After all, I doubt a full size construction will be comparable in price,
effectiveness and scalability to a heliostat field.

Marijan Pollak wrote

--
View this message in context: http://forum.openscad.org/Making-transparent-Object-tp20620p20639.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

I agree that FDM printers will not produce optical lenses.  Resin
printers can come closer, but may need some polishing.  If you can come
up with a single 3D lens object that needs to be printed repeatedly, you
could try to find a resin printer to print a prototype.  I would be
happy to print one if it fits in my printer build volume.

Jon

On 2/28/2017 7:25 AM, Parkinbot wrote:

Hi, ParkingBot,  You are right and You are Wrong.
First thing is size as Dome would be 25m radius, so there is no way to put
sheets on it.
Next I know that it would be expensive to produce  Dome by 3D printer of
ordinary kind.
It would have to have diamond cutter to make surfaces of fresnel perfect,
too.
However there is no need for Optical quality, it would not produce pictures.
If produced as plates, there would be no need for cutting Dome by Laser.
Next this Fresnel would be wrapped inside of Dome, therefore it would not
be flat like
others.
I calculated that One Dome made from recycled plastic or glass cannot cost
more than
100,000 € so 9 of them would not cost more than 2 million with installing
them in place.
Compared to IvanPah, on same land and for same investment there would be 5
times
greater  capacity of Solar with at least 80% of land preserved for food
production.
As there would not be any moving parts, nothing can go wrong or shabby, it
would last
at least 1000 of years, if not forever. It would not need electricity for
own work, too.
No maintenance or repairs lower the cost of electricity produced. If whole
CSP Solar
would cost 10 million $ and last 1000 0f years its cost is 10,000$ per year
or 100,000$
incase it last just 100 years.  With production of 240 MW per day it would
produce
87600 MW per Year, which if sold for 100$ per MW  would pay itself back
allmost in 12
months, and in case it save some factory all cost of electricity, and that
is around 200$
per MW, it would pay itself in one year and save additional 6 milion $.
Add to that Carbon Credits earned and at least million or two for food
produced, so it can
pay itself in 6 months and earn one more in next 6 months.
Add to that 12 to 18 Wind Power Stations of 1 to 4 MW each, able to extract
at least 144
times more energy from same wind, presently used by just ONE standard unit,
You get
another equall sized electricity production. so one would pay itself back
in 3 months and
earn 3 more units, or if electricity is sold at half market price to Grid,
one would pay itself
back in 6 months and earn one more same year, or 2 more each following Year.
Actually since WindSolar would work 24/365, it needs no backup, and it
would produce
20 MW per hour in "High Consumption Periods" when Grid pay Premium prices
for the
electricity received, in US 3.3 times more expensive, in my country 10
times more.
Additiona lincome would pay it back still faster.
Next it can provide electricity for local standard Wind Farms that pull
Gigawats from the
Grid when starting  to work, which my WPSs do not need.
It is therefore Grid independent and can be used anywhere to provide
electricity to local
consumers.  It could be built in groups and groups of groups, and in case
of storing the
energy from other self renewable sources, capacity can be 30 times greater,
producing
on average 300 MW per hour, and capacity of its 18 WPSs can also be 20
times larger,
up to 720 MW per hour total, at cost under 100,000 $ per each additional 4
MW.
That no ordinary CSP Solar can do. Cost of electricity produced drops
linearily with every
MW in capacity added, from initial cost under 5$ per MW from Wind, Where
all of Wind
would be used, from 1m/sec to 100m/sec. and beyond.
I do not calculate with more than 30% od Dome collecting Sunrays, but it
may be more
than that.
In countries like India insolation is 30% greater each sunny day, so total
electricity
produced would be that much larger.
In addition my new Wind turbines would cool ait and so actively lower
temperature in
overheated Athmosphere.
Last but not least, there is process  to capture all gases from Athmosphere
and also
separate them, using electricity, and from CO 2 can valuable Carbon
products, like
most notably Graphene, and that would bring more Carbon Credits and income
to
owners.
If community owned, they would bring prosperity even in most desolate
places, and
at least 40% of population is off Grid and either use very expensive
electricity or are
without any.
Thank You for reading this.

On Tue, Feb 28, 2017 at 1:25 PM, Parkinbot [via OpenSCAD] <
ml-node+s1091067n20639h15@n5.nabble.com> wrote:

--
Regards from Croatia, the Homeland of one of greatest inventors from 19th
and 20th Centuries, Engineer Nikola Tesla!

--
View this message in context: http://forum.openscad.org/Making-transparent-Object-tp20620p20644.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

Jon, on picture it is just 1:100 scalled down "Proof of Concept".
Of scourse standard 3d printers would not be much useable,
and one material that is cheap is recycled glass.
Unfortunately while there is lot of waste plastic, there is no
waste resin that can be reecycled.

On Tue, Feb 28, 2017 at 1:54 PM, jon_bondy [via OpenSCAD] <
ml-node+s1091067n20640h46@n5.nabble.com> wrote:

--
Regards from Croatia, the Homeland of one of greatest inventors from 19th
and 20th Centuries, Engineer Nikola Tesla!

--
View this message in context: http://forum.openscad.org/Making-transparent-Object-tp20620p20648.html
Sent from the OpenSCAD mailing list archive at Nabble.com.