precision quantified
...
and the thing is geared such so that meter gets you through 45 degrees of tilt in the mirror.
to control the mirror to within .25 degree (so that the uncertainty in our mirror is about the same as the uncertainty of our source [the suns angular spread]) we need to know the position of the piston to within ...
45/.25 = 1/x
x = .25/45 = .0055 m
.5 cm
1/5 an inch
whew. so using grey code to read the position you could use marks not quite that big. I'll believe one could make a sensor to read grey code that size.
except, thats assuming everything else is slopless.
really, the piston can only account for a fraction of that. say 70% of the slop is in bearings and gears, piston uncertainty can only be 1/15th of an inch. which means grey code marks 1/16th an inch.
doable?
any other sensing options good to .15 cm over an entire meter?
...
actually, I started this thread to suggest an electric motor geared 64,000:1
but this sounds worth exploring further. how good of an assumption is .25 degrees? or piston uncertainty being a whole 30% of the total uncertainty?
with the electric motor and worm gears i had it all (or half) ready to analyze the linkages and gears and get an idea of what level of machinist would be required. like involute gearing maybe. but what linkages do we use with the linear actuator arrangement?
?
Ozheliguy
sorry I'm not sure who I have the pleasure of replying to?
However I am glad that you have raised these issues. I was mulling
around in my head some similar thoughts trying to get them clear
enough to raise clearly on this forum.
Yes I agree exactly with your comments, I believe we are thinking
along the same lines. With your long 1 mtr travel actuator moving a
long arm attached to your slat. The same principle of gearing/
leverage/ velocity ratios etc. If you are moving your slat with a long
arm attached rigidly to it the end of that arm moves through a very
big arc/ radius. So it takes a lot of input displacement to get a
little bit of rotational angular movement. Even with gearing or
linkage play/ wear etc you can still get your required degree of
accuracy. I cant see how you can move something, ie our mirror without
some form of linkage or gearing. Of course I know there are direct
drive motors and actuators but I dont know of them being any more
precise, or suitable? From a large displacement example continuing
with this theme the automatic roller gate to your local secure estate.
Basically an electric motor driving a pinion with lengths of nylon
'rack' gear attached to the gate. The weight of the gate keep the mesh
resonably tight, reducing backlash. The 'throw'/ displacement of the
mechanisim is only limited be how wide the gate is to attach more
sections of rack gearing to it. The limits are set when the gate runs
in to micro switches at either end of it's intended travel. Sorry
getting a bit side tracked there. Anyway a hydraulic actuator could
work the same way, ie Long arm attached to slat, one pivot connected
direct to actuator output rod. As for grey coded position sensors,
anything is possible I suppose. I have no idea what that sort of
system might cost. I was just think of the photoresistors that other
people seam to be using. I have heard the question of if they can be
accurate enough. I think yes. If they are working backwards, looking
for the sun in a mirrors reflection they will only get that when the
mirror is very accurately aligned. After all it requires, as has been
pointed out to me, a very precise alignment of the mirror to reflect
the sun into the collector. The small sensor, (certainly not 6 inches
the collector is) can be aligned with the centre of the collector
opening aiming at the centre of the mirror we are trying to aim, all
of them individually if that is what is decided on. This has to be
accurate. It is like trying to accurately aim let's say a laser
pointer at a target some distance away. Not easy, every little
movement of your hand is amplified. Now imagine you are the target,
with glases on of course, you could see those large movements and tell
the person with the laser on your cell phone how far off they are or
when they are dead on. Not a good analogy but you see my point.
Percision- Every pilot knows, from his navigation training, that 1 deg
heading error will put him 1 nm off track over 60 nm at the equator.
One degree is not much, ie the smallest graduation on a big compass. 1
nm is more than enough to miss target like an airfield in poor weather
over a short journey (ie 35 mins in a slow plane). This 1 deg is 2 nm
off course for a 120 nm journey, 0.5 nm over 30 nm's etc. A marksman
also knows that is harder to hit targets further away. I'm sure you
get the point.
So with our mirrors it is much easier (ie less angular precision
required) to hit our collector target at closer range. With the
current proposed design our mirrors range between 14.14 and 26.92 feet
from the collector. We can see that it will be harder, (approx twice)
to aim our farthest mirror.
Some thoughts on mirror positions and collector position and the
structure holding everything together:-
Collector- I notice that the UK project has the collector low around
the same level as the mirrors. This has a few advantages- ease of
maintenance access, it doesnt shade any of the mirrors except
partially for the very first bit of sunrise and last of sunset. This
also means when trying to track the sun with sensors it is never
hidden by the collector. The structure is less, the plumbing for the
collector fluid is less. As we all know less structure is less
material and less expense. It gets better still because it is less
effected by wind so can be built lighter and cheaper. Obviously there
is a practical limit to all this. Taking this one step further,
possibly impractably so, it is possible to have the collector below
the horizon ie in a trench?
Mirror positions- The current design as I understand it, is to the
mirrors horizontally, starting 10 horiz feet to the last at 25 feet
from the collector (14.14 and 26.92 feet actual 'slant' distance). If
distances were reduced we would have less structure, less required
accuracy, could have the same structual strength and rigidity with
lighter construction as it has less structure to support itself. In
fact perhaps we wouldnt even require the same rigidity seeing the
accuracy required was less? This also uses less space for the same
power output. As I have also previously mentioned perhaps to go
further with this concept of trying to get a smaller structre and
distance from mirror to collector by reducing the the gaps between the
mirrors. I realise this is to stop/ reduce the mirror shading the
mirror behind it. Perhaps reducing these demensions could allow the
mirror array to be inclined at the back to reduce the issue mirror
spacing and shading? At an inclination of 42 degs above the horizon
the spacing between mirrors could be reduced to less than 2.5 inches.
Or nothing and live with shading some time but get full unshaded
benefit at other times. Of course more windage, but maybe this is
worth it? Just some suggestions/ thoughts to put out there for further
discussion. Getting a bit more radical how about vertical, on the
south side (norther hemeispere) of an existing wall, fence, side of a
house barn etc. Vertical more of a benefit at higher latitudes. Is
also easier to have a lower ground level collector, which is probably
where the steam engine is likey to be. Could be made cheaply if
another structure was taking much of the loads. The frame only needs
to support the mirrors not it's own structure weight. Of course I know
this would be considered ugly, and not allowed by many local councils
etc. Just some more crazy thoughts.
If anyne knows how this achieved currently on existing projects, I
would be interested to hear from you guys. As you guys can probably
tell I dont know so maybe am repeating mistakes that those before me
have made and learned from. Perhaps I'm not limited by traditional
thinking?
Cheers.
Milton.
...
I meant to be specific inorder to generate next steps. I would prefer to chose an idea and pursue it with some rigor until we decide its time to try it or dump it for something else.
your points about the mirror distances are true. I dont think anyone is attached to that specific arrangement. There are reasons for some things and others are accidents. The reciever is in the air to minimise the angle of incidence on the reflectors, which makes them more efficent. But these points are not really important for the question of tracking. It matters not that the closest mirror can hit the reciever accurately if the furthest one misses it.
If we want to talk about tracking, lets pick a section of that topic and explore it.
Precision in the actuator needs to be addressed.
a back of the napkin calculation says a grey code sensor would have to read markings less than 1/16th of an inch. smaller than that is probably necessary. could we even accurately make and read grey code markings that small? is there a better sensing option? if not, the hydraulic and electric linear actuator doesn't work.
if so, then we can think about where bearings and rollers and things will be and calculate how much slop that set up may have.
again, another option is an electric motor geared 80,000:1 (as dwane from red rock suggested to me). If we could count each revolution of the rotor (which is common in control electronics), then we would control mirror tilt to 360/80000 = .005 degrees which is probably orders of magnitude smaller than slop from other areas of the system.
I prefer to follow this topic further. we could discuss optical sensor arrangements or pre-calculated sun paths or reflector/reciever juxtapositions later or atleast seperately.
-Elliot
Ozheliguy
sounds good. Let's get things happening. What is the process, for
progressing the project? Who makes the decisions to proceed, with
what? Who does what tasks? I agree that:-
1. how to move the mirrors is probably the first issue we should be
concentrating on, and
2. how to move them to the correct angle is second.
Although obviously there is overlap with these two.
I'm in the process of getting some price and suitability information
together for the hydraulic actuator option. Ie waiting for email
replies from O ring vendors, tube manufacurers, directional control
vavle vendors etc. When I get enough of that to give everyone some
more information to make decisions on, I will post it.
Do you have an idea of what the costs involved in the 80k:1 geared
motors are? I assume the outputs are attached directly to the end of
the mirror slats and the motor body is attached to the slat support
frame? Is this the case? I heard someone say to me that the gears in
motors were likely to give unacceptable potential play and therefore
not give the required accuracy? I assume that this is the way mirrors
for similar existing arrays are being rotated? I am all for them if
they are cheap enough. I think they would definately be easier than
hydraulic to just buy them. You are thinking of one for each mirror I
take it? Even if there is a failure rate, perhaps this is still
acceptable if the aquistion costs and availability are right. .005 deg
certainly sounds small and precise. Again even though there are gears
with potential for backlash doesnt make them inaccurate. There are
limitations to everything, gear backlash, linkage bearings etc. The
system is designed to allow for those imperfections/ inaccuracies at
their maxium to be well what is acceptable in our system. So even if
using you example of a 80,000:1 gear ratio missing one gear tooth,
which has to be less than 1 motor revolution seeing all gears have
more than 1 tooth, it would still be less than 0.005 deg. Thus still
well within our allowed tolerances.
The same would apply to linear actuators (either elcetric or hyd). An
arm solidly connected to our mirror slats that was long enough 'lever'
to give us the same 80,000:1 if we want. It all depends on relative
lengths either side of our fulcrum ie our ratio. In our case our
fulrum is probably our mirror axis pivot point. Levers are the same
concept as gearing, ie force/ displacement multiplcation, gears and
pulley are just rotary levers. Again we have a bolt and bearing link
attachment ie our inaccuracy potential. Let's say we used a 0.5 inch
bolt at that pivot, even it that bolt was worn to half its diameter
the resulting slop would still be in our system design tolerance we
could have it the same 0.005 deg depending on our arm length.
At the end of the day I think this will all come down to price, and
availability. It can be done as many ways as we can imagine.
Grey/ gray code sensors. I think I understand a little of how they
work, ie used in aircraft altimeters so no friction and reliable when
there is minor corruption they will still work reasoanbly well. As I
understand it they are used to measure rotational angular positions. I
must have missed how they were propsed to be used in this system. As
good a device I believe them to be I cant see why or how in the system
we would use them? I also have no idea of the cost and availability
for these. Sorry to get back to autopilot analogy again. On an
aircraft there are 2 sets of sensors in the system. The first main set
(one for each axis being controlled, and mabe back ups)- these are
gyroscopes (or modern equivalents) that sense where the aircraft is.
Ie nose is down for example. The second set are connected to the
pilots controls so the system knows the pilot has made an input and
not try to correct it, as opposed to a gust of wind it will correct.
In our case all we need is the first set. Ie we just need to know that
the mirror is relecting the sun into the collector, ie photo resistors
are sensing reflected sun rays. We dont need to know what the angle of
the mirror is or what the length or displacement of an actuator
is.
Thanks.
...
Hi Elliot,
sounds good. Let's get things happening. What is the process, for
progressing the project? Who makes the decisions to proceed, with
what? Who does what tasks?
mmm, process you ask?
There isn't one at this point. Other projects that are smaller have been successful with one person taking the lead. But this is so big, I dont think anyone is both ready and able to direct it. I was curious if you had agreed to see a certain section of the process through, because that could be a good method (individuals heading one aspect).
Right now lets just try to sort out what we can, make whatever progress we can, and see where it goes. If the solar turbine never actualises (knock on wood), atleast we'll be better prepared for similar projects that also require solar tracking.
I agree that:-
1. how to move the mirrors is probably the first issue we should be
concentrating on, and
2. how to move them to the correct angle is second.
I'm in the process of getting some price and suitability information
together for the hydraulic actuator option. Ie waiting for email
replies from O ring vendors, tube manufacurers, directional control
vavle vendors etc. When I get enough of that to give everyone some
more information to make decisions on, I will post it.
Do you have an idea of what the costs involved in the 80k:1 geared
motors are?
no. I will look into it and find prices, tolerances, etc.
I assume the outputs are attached directly to the end of
the mirror slats and the motor body is attached to the slat support
frame? Is this the case? I heard someone say to me that the gears in
motors were likely to give unacceptable potential play and therefore
not give the required accuracy?
I was thinking to drive a shaft of worm gears that would move all the mirrors. There will be play in the gears. One just has to quantify that play from tolerances and see if it too much or acceptable.
The same would apply to linear actuators (either elcetric or hyd). An
arm solidly connected to our mirror slats that was long enough 'lever'
to give us the same 80,000:1 if we want.
I dont think so. In a linear actuator you have only the throw of the piston to work with. That throw corresponds to the full swing of the mirrors that you want. You use levers and gears to get that relationship, yes. But at the end of the day you need to know the position of that 3 meter piston to within a couple millimeters.
a spinning motor can spin all day, and the relationship between rotor revolutions and mirror rotation is set arbitrarily by gearing.
At the end of the day I think this will all come down to price, and
availability. It can be done as many ways as we can imagine.
nope, we will imagine many things that won't actually work.
Grey/ gray code sensors.
As
good a device I believe them to be I cant see why or how in the system
we would use them?
I thought you brought them up!
it would be a digital ruler attached to the actuator, with optical sensors reading the currect position.
In our case all we need is the first set. Ie we just need to know that
the mirror is relecting the sun into the collector, ie photo resistors
are sensing reflected sun rays. We dont need to know what the angle of
the mirror is or what the length or displacement of an actuator
is.
Again, i dont think so. Where are these photosensors? right up close to the mirror and they will be too uncertain. Close to the target is ideal except for the heat and...
if a mirror is only 10 feet away, then .5 degrees makes about an inch difference, which is %20 of the light coming from a 6 inch slat. (so, .25 degrees of mirror tilt equals an inch at 10 feet). The sun is 1 degree big in the sky (gives light ith an angular spread of 1 degree). So the photo sensor is saying "good, good" while 20% spills off this side or the other. That assuming absolutely perfect sensing: no internal reflection off the inside of the shading tube giving half signals and such.
If you took your signal from the furthest mirror, I think there will be problems with adjacent mirrors interfering. (at 20 feet, mirrors are only 1.5 degrees apart from perspective of target, and each mirror is giving light with an angular spread of 1 degree.)
but, if some photosensor configuration could work then we do not need to sense actator position. we only need that with pre-programmed sun path calculations or with a dummy heliostat tracking the sun duirectly and giving orders to the rest of the mirrors.
elliot
Ozheliguy
some additional thoughts.
It all boils down to this:
1.Turning the mirrors- I cant see any big problems here, all could
work fine. Elec geared motors, linear elec actuators, or even
hydraulic actuators. I think this will come down to price, convenience
and availability.
2. Accuracy of turning the mirrors- we can 'gear' the amount of
precission into our design easily enough. In the case of geared
motors, that is already conveniently achieved for us. However I can
see that the mirror pivot points, rigidity of the mirror mounting
frame and collector and rigidity of the mirrors themselves (sagging or
twisting etc) will be crucial, they are all on the non geared side of
the system and small dispacements from where they should be may have a
large impact.
Thanks.
> > -Elliot- Ocultar texto das mensagens anteriores -
>
> - Mostrar texto das mensagens anteriores -
Marcin Jakubowski
Hi Elliot,
sounds good. Let's get things happening. What is the process, for
progressing the project?
The process is for people to agree to voluntary action, reviewed by others. We will build the whole system at Factor e Farm after prototypes of components are tested.
Who makes the decisions to proceed, with
what? Who does what tasks?
It is based on initiative. Volunteers agree voluntarily.
Cost is the main point of optimization. This can be done, the question is at what price. Before spending any dollars, I would refer to the other research companies and evaluate their choices.
Marcin
--
To post to this group, send email to solar-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/solar-turbine?hl=en
See http://openfarmtech.org/index.php?title=Solar_Power_Generator_Distillation_Transcript
--
--------------------------------------------------
Marcin Jakubowski, Ph.D.
Open Source Ecology
http://openfarmtech.org/weblog
http://www.replab.org
opensourceecology at gmail dot com
Skype: marcin_ose
--------------------------------------------------
Nobody said that building the world's first open source village would be easy.
-- Anonymous, 2009
A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.
-- Robert A. Heinlein
NOTICE: All discussion in this communication is in the public domain, unless otherwise noted. If you are sharing proprietary, confidential, or otherwise privileged information, you must make that explicit. Otherwise, this discussion may be copied, republished, and otherwise used in the public domain - respectfully and with proper attribution. Furthermore, please consider that we are not interested in discussion as much as action. Therefore, we are particularly interested in discussion of ideas that both parties can commit to by acting on them.
Ozheliguy
I just read your previous replies to my post previous to my last one.
Yes I have agreed to look into some of these issues further,
hydraulics etc.
Perhaps a second hand piston might be cheaper. I will see what I can
find. But our requirements are somewhat uniquie. ie using 400 psi
water, a long throw, slow moving etc I think we can make them cheaply
and easily if it comes to it. 400 psi is not so much.
Yes agreed that an electric linear actuator accuracy would most likely
limited by it's maxium throw as to how much accuracy you gear in.
I cant understand why we need to know what position an actuator is in.
The sun going into the collector doesnt care what position an actuator
is in? We wouldnt need to know how many revolutions a worm gear has
turned. Usually when they get to the end of their intended travel they
run into a micro switch cut off, or similar. I must be missing
something. I would not create more complexity than is absolutely
required.
Yes I tend to agree with your idea of moving all the mirrors using a
worm gear. I dont even think you need a worm gear, a cheaper staight
pinion/spur gear running a rack, with as you say the required amount
of percision designed in. I think we should be able to connect them
all together successfully. There is around a 12 deg angle difference
between the mirrors but they all move the same amount- something less
than 42 degs depending on latitude and time of year.
Yes I did mention grey code sensors, for one of my long winded
analogy. I forget the illustration I was trying to make but never
intended them to position slats. Do you know how other heliostat
systems are aimed? This could save a lot of trying to reinvent that
wheel. I will get some more pictures done. I keep starting some and
realise what problems there are, or how it could work better. Still I
guess that's the point.
Aiming- yes I agree with all your points, well put. I was thinking of
having the sensor up near to the collector. I was thinking on the
vertical collector support using the another reference mirror under
the collector so that it didnt get shaded by the collector. As you say
away from the heat. I aggree that the farthest mirror at nearly 27
feet is the best one to use as the aiming mirror. But there is a
couple of times each day when that mirror gets shadowed by the
collector as the sun rises and falls through 21.8 degs elevation each
day. Another thought I had was to have a small reference mirror,
attached to but extending past the longest slat. The smaller mirror
(or could mask off the 6 inch mirror except for a small strip in the
centre ie 3 inch) would shoot a smaller beam giving the photo
resistors a sharper point to pick up. Good point also about possible
interference from the other slats if using the longest mirror. Perhaps
we need some sort of shield. Perhaps a PVC tube?
Thanks.
On 28 fev, 23:09, "..." <offonoffoffon...@gmail.com> wrote:
...
But our requirements are somewhat uniquie. ie using 400 psi
water,
why water?
I aggree that the farthest mirror at nearly 27
feet is the best one to use as the aiming mirror. But there is a
couple of times each day when that mirror gets shadowed by the
collector as the sun rises and falls through 21.8 degs elevation each
day. Another thought I had was to have a small reference mirror,
attached to but extending past the longest slat. The smaller mirror
(or could mask off the 6 inch mirror except for a small strip in the
centre ie 3 inch) would shoot a smaller beam giving the photo
resistors a sharper point to pick up. Good point also about possible
interference from the other slats if using the longest mirror. Perhaps
we need some sort of shield. Perhaps a PVC tube?
mmm. this needs more thought, but a mirror pointed at a seperate dummy target is a good idea. could not be on the same line as the actual target. Also, the sun moves from maybe 25 degrees above the horizon to past vertical to being in the north in the summer at noon, so any mirror would be shaded at some point.
also, I was wrong about solar size. It is .5 degrees. still, that ammounts to the furthest mirror mostly missing the target (reflection moves 2 inches on a screen perpendicular to the light rays, which would be more than 2 in on a tilted target). Even with perfect shading in an infinitely long pvc pipe of almost zero diameter that doesn't have internal reflection. any shade that is shorter, wider or has some internal light diffusion will do worse than this.
To me that rules out using a sensor at the target. what do you think?
though adding a linear lens to the fake target could make it work...
Ozheliguy
need to create another system- keeping things simple, cheap etc. This
was one of the reasons I was suggesting hydraulic actuators. For
electric actuators we will need an electric system as well, ie
batteries at the least, more Complexity, expense etc.. This of course
could be recharged, or run when a generator was online. As you say
though having the solar collector system independant of the electrical
generation system is preferable. An effective solar collection system
is still good even if the steam engine electrical power generation
cant be made a reality, or it takes a while etc. I have many uses for
hot water. Apart from the obvious showers/ kitchen, house heating,
thermally driven chillers/ refrigeration airconditioners, atmospheric
water generators and of course as mentioned electrical power
generation. The other reason was reliability of electrical components
largely exposed to the weather.
Speaking of electrical generators, I read the page on selection of the
steam engine. I understand the project was originally looking to use a
Tesla turbine. I couldnt find the reason this was changed. I think I
heard someone say reliability was questionable? Can anyone elaborate
on this for me please?
Yes, I agree about the problem of the collector shading the mirrors at
some point. This is one of the reasons I was suggesting a different
position for the collector. But I also accept that it could be a loos
of efficiency with greater angles of incidence and reflection. Do you
know what sort of losses are involved with this? I dont think the sun
ever rises past the verticle does it? I thought it starts and ends at
0 deg elevation, ie at the horizon at sunrise and sunset. Then rises
to its max elevation, always less than 90 degs, but varies depending
on latitude and time of the year. I am not an expert though, I could
be wrong.
I'm not sure what you mean by a linear lens. Yes I was thinking about
reflections in the tube, painting matt black etc. Maybe it wont work?
It is the best I can come up with so far, but certainly open to input
from those that know more about this stuff than me. Someone out there
must know how this gets done on other existing working projects? I
also agree that we should perhaps have a wider collector opening to
allow for some of these limitations inaccuracies which are a fact of
life. If it was say 10 inches, 2 inches either side of 6, so that any
light within that opening is still directed to the collector/
receiver/ target tube. I dont know too much about this stuff- is this
feasable?
Thanks.
...
Water, we have it already, pressurised. in our collector so we dont
need to create another system- keeping things simple, cheap etc. This
was one of the reasons I was suggesting hydraulic actuators. For
electric actuators we will need an electric system as well, ie
batteries at the least, more Complexity, expense etc..
mmm. I assume you have thought about the corrosion issue using 350 C water.
Also, you too need a battery.
I have many uses for
hot water. Apart from the obvious showers/ kitchen, house heating,
thermally driven chillers/ refrigeration airconditioners, atmospheric
water generators and of course as mentioned electrical power
generation.
Yes, I agree about the problem of the collector shading the mirrors at
some point. This is one of the reasons I was suggesting a different
position for the collector. But I also accept that it could be a loos
of efficiency with greater angles of incidence and reflection. Do you
know what sort of losses are involved with this?
eff = Cos(2*theta_incidence)
I dont think the sun
ever rises past the verticle does it? I thought it starts and ends at
0 deg elevation, ie at the horizon at sunrise and sunset. Then rises
to its max elevation, always less than 90 degs, but varies depending
on latitude and time of the year. I am not an expert though, I could
be wrong.
At the equator the noon sun is just as often north of vertical as south. google "sun path". Even north of the tropics, the sun can still rise in the north in the summer.
I
also agree that we should perhaps have a wider collector opening to
allow for some of these limitations inaccuracies which are a fact of
life. If it was say 10 inches, 2 inches either side of 6, so that any
light within that opening is still directed to the collector/
receiver/ target tube. I dont know too much about this stuff- is this
feasable?
Having a high concentration ratio (width*number of slats vs target size) is important for generating electricity. But less accuracy is needed for direct thermal use.
This makes me think of a less precise approach to produce lower pressure steam for use in all the wonderful things lower pressure steam is valuable for. Such as two types of chilling, all sorts of adsorption and chemical processes and also preheating combustion gas. but just because less accuracy is needed, that doesn't mean that a total lack of precision will work.
I look forward to what drawings you come up with, and what solutions to corrosion and storage of hydraulic motor force.
-elliot
Ozheliguy
Firstly sorry for the delay! Secondly I have downloaded pictures
scanned in somewhat of a hurry so please excuse them being upside down
etc.
Mirror aiming- arm 1 metre long, solidly attached to mirror holders.
Second arm scissor through mirror pivot point with aiming adjustment
screw at opposite end to mirror. Hanging vertical in mid arc travel.
If space restricted can have vertical up, or even horizontal arms for
a vertical mirror array. May need counter balance weights opposite
arms to ease effort required by actuator. Common rail attaches all
arms together, spacing same as mirror spacing. 17 mm travel/ 1 deg of
mirror rotation. This is the crux of a cheap reliable, very precise
mirror rotation system- a long arm and a long actuator through.
Expensive, high precision actuators are not required.
Hydraulic- The main problem is the potential heat of the fluid
involved 400F/ 250C. We would need to cool ‘hyd’ fluid/ water using
collector inlet water. Possibly regulating the pressure lower to
something like 100 psi might be better too? It could be easier to just
thread the tube ends and screw caps on? Does anyone know of a
reasonable supplier for hydraulic, double solenoid, directional
controller valves? Stainless steel tubing something like
http://www.mcmaster.com/#mil-t-8808b-stainless-steel-tubing/=65asoo.
I have thought of corrosion previously. Corrosion is one area that
unfortunately has caused me a lot of work, in the past so I am
particularly wary of it.
Electric- Rack something like ( http://www.mcmaster.com/#rack-and-pinion-gears/=659zpi
product 5170T1 ) and a small pinion drive, 12V stepper motor 1.8 deg,
around $30, (200 steps/ revln), or 900:1 geared. Also need stepper
controller. Each step equals = a fraction of a degree mirror rotation/
or moves reflection on collector of furthest mirror a fraction of a
mm. I don’t know what the prices are but going the whole hog we could
use a 15,000:1 geared stepper motor like http://www.mclennan.co.uk/product/gearedmotorassy.html.
or http://www.mclennan.co.uk/datasheets/european/geared/p5stseriesgearedsteppermotors.pdf
This would give 720,000 steps per revolution. Getting a little off
track, but I did come across an interesting optical rotary position
sensor supplier http://usdigital.com/
Aiming- separate small/long but narrow, aiming mirror. Ideally the
sensor fixed, facing away from sun direction. Long way away better.
Have to make sure doesn’t pick reflection off other or multiple
mirrors reflections. Aim sun spread equal distance from centre of
collector target. This mirror could even be placed in front of the
collector slightly, ie opposite side to the other mirrors. This may be
required for sun elevations past the verticle- ie in the tropics.
.
Speaking of steam ejector cooling, this is one of my interests at the
moment- do you have any information on it?
Concentration ratios- I am sure you are correct about required
concentration ratios. I wasn’t suggesting to dilute/ lessen the ratio.
I think we could still have the collector/ target tube the same
diameter but increase the parabolic concentrator opening size so that
even if it was not catching light evenly across the whole surface.
Anything it did catch would get directed onto the target. I would
imagine this to be the case- is it not?
Cheers.
On Mar 8, 3:52 am, "..." <offonoffoffon...@gmail.com> wrote:
Marcin Jakubowski
Marcin
--
To post to this group, send email to solar-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/solar-turbine?hl=en
See http://openfarmtech.org/index.php?title=Solar_Power_Generator_Distillation_Transcript
Ozheliguy
the pictures are loaded as a file on this dicussion home page called
"Solar_Mirror_aiming_system" Sorry I may have put them in the wrong
place? If there is a more appropriate place for them please let me
know.
Thanks.
On Mar 10, 1:30 am, Marcin Jakubowski <joseph.dolit...@gmail.com>
wrote:
> >http://www.mclennan.co.uk/datasheets/european/geared/p5stseriesgeared...
> > This would give 720,000 steps per revolution. Getting a little off
> > track, but I did come across an interesting optical rotary position
> > sensor supplierhttp://usdigital.com/
> >http://openfarmtech.org/index.php?title=Solar_Power_Generator_Distill...
>
> --
> --------------------------------------------------
> Marcin Jakubowski, Ph.D.
> Open Source Ecologyhttp://openfarmtech.org/webloghttp://www.replab.org
> opensourceecology at gmail dot com
> Skype: marcin_ose
> --------------------------------------------------
>
> Nobody said that building the world's first open source village would be
> easy.
>
> -- Anonymous, 2009
>
> A human being should be able to change a diaper, plan an invasion, butcher a
> hog, conn a ship, design a building, write a sonnet, balance accounts, build
> a wall, set a bone, comfort the dying, take orders, give orders, cooperate,
> act alone, solve equations, analyze a new problem, pitch manure, program a
> computer, cook a tasty meal, fight efficiently, die gallantly.
> Specialization is for insects.
>
> -- Robert A. Heinlein
>
> NOTICE: All discussion in this communication is in the public domain, unless
> otherwise noted. If you are sharing proprietary, confidential, or otherwise
> privileged information, you must make that explicit. Otherwise, this
> discussion may be copied, republished, and otherwise used in the public
> domain - respectfully and with proper attribution. Furthermore, please
> consider that we are not interested in discussion as much as action.
> Therefore, we are particularly interested in discussion of ideas that both
> parties can commit to by acting on them.- Hide quoted text -
>
> - Show quoted text -
...
In case it matters, the elevation of the summer sun is higher than the noon sun throughout the day. and, likewise, the winter sun is lower than the winter noon sun over winter days. so, there is a broader range of angles to consider.
The aiming mirror where you place it would be shaded in the mornings and afternoons in summer. But it could be placed even further south and still do its job. I still am not convinced that this method, as you describe it, can be accurate enough. It can probably be adapted. But it has a minimal slop of .5 degrees (.25 degrees mirror tilt). .0087 radians * 25 feet = .22 feet or 2.64 inches of play where the rays from the last mirror hit the target. This is 2.64 inches of arc length, which would be a greater distance projected onto the plane of our target, which is not pointed directly at the last mirror but towards the center one. so, atleast 3 inches of play from strictly geometric arguments. Which is not too bad, but combine that with our limits of precision in mounting the sensor, and the sensor picking up reflections inside the tube... But something like that is probably workable.
The target size determines the concentration ratio. If you make it bigger, you decrease the concentration ratio because you give the target more area from which to radiate heat. given a target size, the precision determines your optical efficency and also your effective concentration ratio.
Corrosion! stainless steel will corode. If you are using presurized water, you cannot cool it below the boiling temperature without causing steam to keep condensing into the water, raising its temp. All the trouble to boil water under pressure just to keep gainlessly recondensing some of it, what a waste of energy that would be. I think using water is a losing idea also because you cannot take this to a hydraulic machinest to get a quote on how much it would cost for parts, labor etc. You will get laughed at or atleast ignored.
Steam ejectors are pretty simple. What more information do you need? Beyond the principals, which you probably understand, there is only intricate engineering equations and fluid dynamics. I don't know enough fluid dynamics to design one.
Ozheliguy
my appologies for upside down pictures. I will have to try and learn
how to drive the work scanner, when they are not so busy with it.
Sun angles. Yes I freely admit I am not an expert in this area. What I
used for my drawings was a sun chart for where I live at 7 deg S
(resonably tropical) and another at a location 45 deg N (reasonably
high latitude) for a general comparison of extremes of where the
majority people would live. Yes I agree with you that forward (further
towards the sun from the collector) could well be a better position
for the aiming mirror, to allow for higher tropical sun angles. I
suggeded this in one of my earlier posts.
I also agree with you that the sensor position would be better further
from the aiming mirror to give greater accuracy. Where I have drawn it
on the collector support it might be something like 9 feet from the
aiming mirror. When the farthest mirror is alot further away, 42 feet
I think. I am not sure where it could be better put without running
into issues like extra structure, more space, and sensing the wrong
reflections. Any suggestions? Ideally I think it would be better
being: narrower than 6 inches perhaps say 1 inch wide (effectively
making it 6 times as sensitive?), further away from the target mirror
reflecting the sun, facing the opposite direction to the sun and not
likely to pick up other reflections. But I cant think of how to do
that, yet. Anyone with a better suggestion is welcome to dazzle us.
Again any ideas how this is currently achieved on current functioning
solar tracking mirrors, or how we can find out? They might have
already solved all of this stuff.
Target size. I was not suggesting to alter the size of the collector
tube. I was asking if it was possible to have a parabolic concentrator
with a larger opening. I dont have the same level of understanding
with this as you and maybe it effectively dilutes the concentration
onto the collector. It would be good it this were not the case to
allow for slight system inaccuracies, misalignments etc.
Corrosion. Yes you are also correct that Stainless steel can corrode.
It all comes down to the grade of the Stainless steel and how
corrosive the fluid is. Hot water wont corrode the grade I mentioned
in my drawing. It will most likely corrode lesser grades. We have to
keep all of this in perspective we have had cast iron hydraulic wheel
cylinders in our car brakes for many years. Using a fluid that absorbs
water. They are far more critical for our well being. We are also
talking about a steam engine made from mild steel and cast iron.
Yes perhaps the hydraulic idea isnt workable? I was thinking of a
pressure reducer that would lower the boiling point, and having the
cooling water recycled (now pre heated) back into the inlet of the the
collector. Maybe that is getting a too complicated. I am looking into
some other ways of doing it, but let me get some more info first.
These are all just ideas to be looked at, at this point. I still think
it will come down to the cost for what is capable of doing the job. We
are just trying to come up with some possibilities that we can wittle
down. Much better than no ideas. So what is your current thinking on
what is workable. I will chase more details of what is agreed on.
Geared stepper motors, driving a rack, or something else? We still
have to cost all of these other alternatives yet, which is why I am
keeping an open mind. If we limit our thinking to the accepted ways
things are done now- we can expect the same results. That is a small
scale solar collection and power generation for a resonable afordable
price- NOT yet achieved.
Steam ejectors. I know how they work. Any information you might have
about using them on a small scale for airconditioning around say 10kW
cooling capacity. I have seen some research feasability projects at
this size, or slightly smaller (5kW) being driven by parabolic solar
collectors. I can send you what I have if you like? Everything else I
can find, ie currently produced, is all much larger scale. Up until
this sort of project is made viable/ workable homes haven't had a
supply of steam to use for such purposes. Industrial factory plants
and old Steam trains have been the only users of this technology.
Firstly information on how this can be done on a small scale and
secondly where condensors, ejectors and vacuum evaporators can be
sourced or made. They are after all reasonably simple devices- 1 Atm
press capable containers, made mostly from steel. Anything would be
greatly appreciated. I'm sure you can see where I am coming from with
this. Home airconditioning with almost no moving parts, no compressor
vibrations, wear or current draw, no CFC's or any other chemical other
than water, works best when it is needed most- ie when the sun is
shinning. Just think how much less electrical capacity homes could
require if ac wasn't a load. 3kW Steam powered steam engine generator
would be adequate for a lot more homes. Of course all this once again
also hinges on our ability to generate adequate steam at a reasonable
cost, which is the task at hand.
What do you want me to look into next?
Cheers.
Ozheliguy
A couple of thoughts I had recently.
I have loaded another 2 more drawing into the docs section called
“Solar parabolic concentrator sizes” and “Thermal \hydraulic solar
tracking device”.
The parabolic concentrator around the target/ collector tube. I
realise that the concentration ratio is a ratio of collector area to
target area. In this case 16 mirrors X 0.5 ft each is a total of 8
feet/ 96 inches. Our target collector tube is 2 inch diameter.
Therefore our concentration ratio is 96/2 = 48:1. I am not suggesting
changing these dimensions. I would think that, in my very limited
knowledge of this subject, that increasing the opening size of the
parabolic concentrator should allow for some misalignment/ inaccuracy
and still catch the same amount of energy? Perhaps I am over
simplifying this issue.
Aligning the mirrors using a ‘target’ mirror. Some doubt came into my
mind if this will actually work? The sensor/ probes are looking for
the sun. If it is all aligned correctly the probes see a reflection of
the sun in the mirror. But if the mirror is not aligned the reflection
would be bounced back behind, and to the side of the sensor- way out
of the sensors field of view. It may track the reflected image to one
side. This may depend on where the mirror rotation axis is- ie on the
surface of the mirror or behind the mirror. If behind the mirror
surface will move in an arc, not just pivot. If using a target/
aiming mirror doesn’t work it may be possible to use some sort of
shroud/ lens/ reflector arrangement to make it work as desired. I have
to think some more about this. I have no practical experience with
these devices so for those of you out there that do know more about
these things all suggestions gratefully accepted.
Cheers.
> > fluid dynamics to design one.- Hide quoted text -
...
I would think that, in my very limited
knowledge of this subject, that increasing the opening size of the
parabolic concentrator should allow for some misalignment/ inaccuracy
and still catch the same amount of energy? Perhaps I am over
simplifying this issue.
Are you talking about a Coumpound parabolic secondary reflector? It has a definite and fixed ratio between its entry and exit aperture based on the angular spread of light coming in (distance between first and last mirror slats). So, you cant just make the primary reflectors innacurate and make a secondary reflector that makes up for it. What the secondary can do depends on what the primariy gives it.
If you could just make the secondary aperture bigger, then why did we have it it the size it was?
you can catch the same amount of energy, but at a decreased concentration ratio. That is, with slightly less useful energy.
...
What do you want me to look into next?
you don't have to take orders from me, but I would like to see a document that summarizes and compares the costs of these different options.
A system with a light sensor vs a system that uses precalculated sun positions, and hydraulic vs electric power systems.
This would include:
All the significant parts each system would need.
Which parts are unique to that system.
what are the monetary costs of those components (even just ball park guesses).
predicted accuracy.
estimated system performance with that accuracy (expressed as kilowatt hours per year or something).
durability issues.
other issues.
Looked like you had some of this information already, so just laying things out side by side would be helpful.
I would be happy to take care of the 4th and 5th items (precision) after I know the details of systems the parts are for.
This could be working document, which we review, discuss and modify.
thats what I would like to see.
-elliot