How Best to charge 12V battery from Solyndra solar tubes?

Discussion in 'The Projects Forum' started by madmadscientist, Feb 21, 2013.

  1. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
    I saw an interesting ad on Craig's List the other day.....
    Does anybody think I have a 'Craig's List problem'?
    I can quit anytime I swear! ​

    Anywho, the ad was for the individual solar cells for a Solyndra solar panel.
    Yes that Solyndra, the failed clean tech company that the repubs have made so
    much political hay over.
    The price was right, less than $8 each so I grabbed a bunch.​

    Check em out

    [​IMG]

    Don't they just look cool?​

    Closer-up view of the tubes.

    [​IMG]

    They are a dark, dark purple with a silver spiral wound around the inside.
    In person they are quite striking. They have the same form factor as a FL tube.
    They are about 3' long and 1" in diameter. ​

    Thought I'd run some tests in direct sunlight.​

    [​IMG]


    Single one up close on a blue background. ​

    [​IMG]



    I picked up the tube and held it at a roughly 45 deg angle so that light could hit the tube from all sides the readings changed to this.
    Open Circuit Voltage

    [​IMG]


    Amps

    [​IMG]

    So I've got 120VDC and .040A. I supposed this supports Solyndras assertion that these tubes are omni-directional.
    I will need to test these under a load but I'm not sure what a safe load would be....anybody got any ideas? ​

    My Crazy Plan
    These tubes are so striking that I want to use them in some sort of outdoor art piece.
    Maybe the tubes charge a battery during the day and then the piece lights up at night.
    Or it could be a kinetic sculpture that moves during the day when the tubes are producing power.​

    Problems
    I don't instantly know how to make use of 100-120V DC.
    I'm not sure that DC-DC step down converters exist that can handle this high of an input voltage.
    I need to figure out how to charge a battery from this input.​

    Anybody got any ideas? ​
     
  2. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    I like the concept. So you got 13 of these?

    120V at 0.4A is 4.8 watts, and with 13 of these your total is 62.4 watts. That would be the maximum power available to run the sculpture.

    Yeah, of course you can convert the voltage to something higher or lower, whatever is needed. A switching converter will do this with very good efficiency.
     
  3. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
    Hi Ernie and thanks for replying.
    I haven't completely set in stone the form of the sculpture.
    Actually, I'd like to do a couple of different sculptures.
    I would like for one to charge the 12V battery during the day and then 'come on' at night for several hours. I would like to power LED's most probably.

    Another idea is for a kinetic sculpture that would be directly powered by the tubes. I suppose for this I would need a DC motor that could take the input directly.

    Do you have any suggestions for a switching converter?

    Yes I have 13 of these.
    thanks,
    Mads
     
  4. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
    Okay well maybe not exactly in the spirit of this board.
    I've found several components at Jameco that I think I can put together to make this happen.
    A switching power supply that is 78% eff.
    A solar charge controller with several interesting features.
    It turns itself off at night, but you can set it to stay on for a set amount of hours to power something.
    And of course a simple sealed lead acid battery.

    The output of the solar charge contoller is on whenever the panels are generating electricity. I don't want the LEDs on this project lit during the day.

    Now here comes the 'roll my own' part.
    I think what I need is a simple photo-switch on the output of the charger.
    The switch will only turn on when its dark.

    Does anybody see any logic bombs with the above or any parts incompatibility?
    thanks so much,
    Mads
     
  5. #12

    Expert

    Nov 30, 2010
    16,247
    6,744
    It's called a, "Dark Detector". I saw one on this site a few days ago. The advice was, "google, 'dark detector, relay'."
     
    madmadscientist likes this.
  6. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
    Well, I screwed up and used the wrong power supply.
    The I'm using is only good for .9A
    I'm going to switch it out for a 4.2A version.

    I've decided that I want to use yellow flicker LEDS.
    Does anybody have a prefered source to buy these in bulk?
    I would like the LED to be as large as possible and the color and flicker to be
    as realistic as possible.

    Thanks,
    Mads
     
  7. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    395
    Might use small yellow flicker to drive 1W yellow LED [ Cat # LED-241, All Electronics ] Not sure where I put invoice but believe $ 3.50 give or take. Ckt may be in Bill's Blog.
     
    madmadscientist likes this.
  8. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    395
    Might check Completed Projects, wayneh, 11-8-2011
     
  9. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0




    Proof of Concept-Engineering Prototype

    Check out my super hi-tec experimenters set-up below.
    [​IMG]
    Yep, that's a shipping pallet and saw horses.

    For now I've decided that I will use 8 tubes. Four pairs in parallel of two tubes in series.

    Open circuit that gives me.

    Volts
    [​IMG]

    milli-Amps
    [​IMG]
    So with P=IV that's a whopping....25.86Watts...hmmmmm

    I've recently learned that switching power supplies can take DC input? Who knew?
    Ordered a 15V output unit from Jameco.
    Here it is hooked up to the array and generating a useful voltage.
    [​IMG]

    When the array is loaded down with just the power supply the voltage drops to.
    [​IMG]
    221V
    The power supply is connected to a nifty solar charge controller, also from Jameco.
    [​IMG]
    Besides the manual being written in some fierce Chinglish its pretty awesome.
    It monitors the battery so it won't overcharge, or undercharge. When the battery is fully charged
    it converts to a float charger. You can also power something off the controller while the battery is charging. It has a built-in timer which is going to be really handy for this crazy idea.
    Once the sun goes down you can set it to power the output for a set amount of time!
    The only problem with this output is that it appears to be 'on' all day also...I don't want the lights
    on the art piece to be on all day only for several hours after it gets dark.

    That grey thing next to the SCC is my solution to the always on problem.
    Its a photocell switch powered by 12V. It turns on when it gets dark and allows current to flow
    to the load.

    The load
    Just for now I have the set up powering an old bit of El-wire. It glows very brightly during the night.

    One thing I have to say about these Solyndra tubes...they are SUPER ITALIAN! As in VERY FRAGILE!
    I've already broken two....
    So know I'm just going to let this thing run for a couple of days and see what happens. I don't have a lot of confidence that I have the controls set correctly on the SCC....stupid Chinglish.

    Questions:
    What kind of electrical connectors are those and where can I get some?

    I'm not sure how many pairs of tubes I'll need to pull this off. I'm not sure how a switching power supply works. I believe that some DC-DC converters take a higher voltage at a certain current and can down convert it into a lower voltage with a higher current. Is that's whats happening here? If not, I don't see how I'm going to get any useful power from 111mA...

    When I hook up the SCC to the PS the array voltage swings all over the place...could be a sign of not enough current? Would it be possible to connect some 'boost caps' to the PS input?

    And finally, what in the heck am I going to do with these things after I solve the electrical engineering problems???

    Mads ​
     
  10. #12

    Expert

    Nov 30, 2010
    16,247
    6,744
    25 watts beats a lot of alleged projects that get posted here.

    Which connectors?

    I think, "boost caps" is a false idea, but capacitors can be used to smooth out changes and they don't waste any energy.
     
  11. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
    Sorry the above was a cut/paste from my blog and the stupid 6 picture limit here (why does it matter when the pics are hosted elsewhere???) makes me edit them out.

    Here's a shot of the connectors, from the two broken tubes...
    [​IMG]

    I'm probably being sloppy with my terminology. I know the car stereo guys use huge 2 Farad caps to make up for voltage dips when the bass hits really hard.
    So, is the same thing possible to do here on the output of the solar array to keep the voltage above 127V which is the lower min for the power supply.
    thanks for the reply
    Mads
     
  12. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0


    Solyndra Art Project Part 3



    The one where I go, 'How many of these freaking tubes is it going to take?'

    Okay, its a bright sunny day, I added two more tubes for a total of ten tubes, five sets of 2 in a series parallel combination.

    Short Circuit Current
    [​IMG]
    Looks pretty good. With the open circuit voltage of 234V that gives me a power of;
    P=IV = (234V)(.143A)=33.46W not bad I'm thinking.

    With the PS hooked up but no load the numbers look like this.
    [​IMG]
    That's 15.17V on the output of the PS and the array voltage drops to 229VDC.

    In the same situation the current is
    [​IMG]
    .0118A Does this itty-bitty current make sense?

    If I hook up the SCC the numbers do this....
    [​IMG]
    That's not good I think...The current jumps up to .152A and the array voltage drops to 52VDC.

    At the same time the output of the PS is.
    [​IMG]
    What I think this means is that the PS is off and I'm just reading the battery voltage. The power supply has a low VDC cutoff of 127V anything less than this and it shuts off I believe.

    To test that somehow the PS was still working at ~50V I unplugged the array from the PS and plugged it into AC mains and got this.
    [​IMG]
    This is with the SCC hooked up...so the PS has no problem running the SCC when its hooked up to AC.

    I think this means that the ten tubes are not enough and can not supply enough current to run the SCC. I realized too late that I should of tested the current running to the SCC when the PS was powered by the AC mains. Then I would know how much power it was using.

    How Many Freaking Tubes Am I Going To Need?
    Apparently ten is not enough...

    I disconnected the SCC from the PS and plugged in a 12V 20W halogen light. It would not stay lit. It pulsed. Does anybody have any idea why?

    The PS is 75% efficient which means it should have (.75)(33.46W)=25.09W shouldn't that be enough to power a 20W light?

    Though, I believe that switching power supplies don't like certain types of loads? When I hooked up a switching power supply to a windshield wiper motor the same thing happened. It pulsed the motor and would not stay on.

    I would love any pointers in figuring out how many tubes I'm going to need to do this. It better not be like 20....maybe this is why Solyndra went out of business?

    Now I realized the tubes are flat on a table so possible not capturing as much sunlight as possible but still....Arrrrrrgh!
     
  13. #12

    Expert

    Nov 30, 2010
    16,247
    6,744
    You're running into the cold filament thing. Incandescent lights take about 10 time as much current to start as they do to run. So do electric motors.
     
  14. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
    Update
    I just realized that there is a logic bomb in my above analysis of the data.
    It would be a impossibly bad Solar Charge Controller that let the battery back feed the solar array.
    I believe that the 12.65V output of the PS is actually what the PS is outputting.
    That the 120VDC min is for the PS to output its rated power.
    So at 50V input it outputs 12.65V.
    This is slightly encouraging as now I can just add pairs of tubes until the output is say 14V, that should be enough to keep the battery charged no?
    Still have to figure out how many tubes I will need to do this.....

    Mads
     
  15. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
  16. patricktoday

    Member

    Feb 12, 2013
    157
    42
    Wow, you're getting some pretty respectable results out of this so far. The tubes have a nice sheen to them as well. Until Jameco opens up its own discount Solyndra solar tube section and it all goes mainstream it will be up to you to demonstrate what on earth to do with those things ;)
     
  17. madmadscientist

    Thread Starter New Member

    Feb 21, 2013
    10
    0
    Saturday, March 16, 2013

    Solyndra Art Project Part 5



    The one where I say, 'How many dang parts are there going to be in this series?'

    FINALLY bought some more Solyndra tubes! As it turns out, the guy I'm buying them from also bought a ton of spare tubes. He is in fact, not taking apart panels to sell the tubes...Which means I'm going to have to figure out how to safely and weatherproof-ly mechanically and electrically connect to these fragile things....

    But first, another exciting shot of my home made Solyndra tube array now with 4 more tubes!

    [​IMG]
    Scintillating isn't it? What is exciting is that the battery was so discharged after several overcast days that the SCC would not turn on the load at night. But today, a bright sunny day, by 1pm the SCC was reading the battery as fully charged! Yahoo!!!

    Here's what the outputs from the array and Power Supply(PS) were showing.
    [​IMG]
    That 13.71V output from the PS is sure better than the previous best of 12.6V.
    Yet the array is still being pulled down to 58V?

    Lets check the current
    [​IMG]
    .335A not bad the array is outputting (.335A)(58V)=19.43W that's an increase from the previous reading of 12.6W.

    But what's the Power Supply output?
    [​IMG]
    Wow, 1.36A and 14.99V, that's a power of 20.34W!!!
    So, apparently I've discovered a 'over-unity' device....Because somehow the output of the PS is more than the output of the array, even though the PS is only ~78% efficient...

    Gotta be measurement error. The black DVM is a super cheapo model.

    What's It Mean?
    I think it means that 14 tubes is the magic number to keep a 12V 7.5AH battery charged even with a ridiculously large load....of course this is in this 'most ideal' setting.....
    If these tubes get used in an art piece they will be a part of the sculpture...not sitting off to the side in a rack...that means much less output from the array.

    Of course, I will not be having such a ridiculously large night time load on the sculpture....So, maybe I can get away with 14 tubes..but I think I'll up it to 16 and call it good.

    Up Next
    Sketches of ideas for what I'm going to do with all this stuff.
     
  18. ronph

    New Member

    Feb 4, 2013
    25
    1
    Hey MMS, I have seen this solyndra tubes being sold at $9 a pop and seems to be fragile alright. Weight to power output also seems to be an issue unless of course its for commercial purposes but for say hobbyist trying to make something out of this tubes would be proportionally large as you are now getting results :eek:.

    Anyway, when you have completed the project, give us your best estimate of the weight e.g. tube rack + tubes combined.
     
  19. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,170
    395
    Instead of a flat reflector, a curved one will reflect light from a wide angle on to a tube. The curve is derived by wraping a string around cylender, attach a pen to the string & starting with pen at surface, draw a curve as string is unwound. This was used for a solar air heater using "tin cans" soldered together for tube, hot wire cut foam for troughs, aluminized adhesive backed plastic for reflector.
     
  20. EkriirkE

    New Member

    May 9, 2013
    1
    0
    Sorry to bump, but just throwing in that I, too have obtained a bundle of these for $8/pop. Re: using an AC switching supply and using the DC input: You can probably squeeze a hair more efficiency ou bypassing the bridge rectifier at the AC input. Solder your solar DC output directly to the BR's DC output (check polarity). Also building a 12V SLA charging setup :)
     
Loading...