REPLACE solar panel with DC supply

Discussion in 'The Projects Forum' started by johnmhome, May 3, 2012.

  1. johnmhome

    Thread Starter New Member

    Apr 19, 2012
    4
    0
    I'm in the odd position of wanting to REPLACE a solar panel with a DC charger supply. Plenty of projects here to go the other way round, but I assure you I'm not crazy!

    Let me explain what I need to solve:

    I have 2 outdoor lights that have motion sensors and are in locations were we cannot get 110v power to them easily. I bought some LED lights that are very bright (10W or so) and require no external power. They have a 6V lead acid cell inside and a solar panel that trickle charges the cell. When the motion sensor activates the LEDs draw power from the battery, but that's not very often. They work great, EXCEPT.....they don't get enough sun on the solar panel due to the very heavily wooded environment of the house, especially in winter. Thus the battery does not recharge enough and eventually no more lights.

    By an odd twist of fate, I have a low voltage source nearby both of them. One is 16V AC, the other is 12V DC 0.5A. My thought was to use these low voltage sources as a constant trickle charger for the lead acid cell, and dispense with the solar panel. The solar cell outputs 13v with around 300mw in full sun.

    Let's start with the AC source. It's coming off the doorbell transformer and it lights up the house number, which is directly below one of the lights. I figured I could put this through a bridge rectifier and capactitor, and then a 12v regulator. I figured the current draw would be very low so I was not too concerned about heat dissipation.

    WRONG! It turns out the lamp has a regulator inside to drop the 13v from the solar panel to the lead acid cell charging voltage of 6.8v. The regulator looks like a zener diode and it pretty quickly got hot enough to melt the solder. Yikes.

    In retrospect I realize the lamp circuit is using the solar panel to limit the max current, so a simple zener diode will usually work just fine but clearly I can't just stick 12V 1A across this.

    The lamp circuit seems undamaged by the hot zener diode - the solar panel will still put 6.8v at the battery terminals to charge it.


    So here's my challenge: I have 16V AC from the house number sign. I can rectify that but how do I then get 10-12V DC with a max current of, say, 200ma? No matter how much the zener diode attempts to draw, I need to limit what it gets. I know enough about electronics to solve all the problems except this. It's at times like this I regret abandoning hobby electronics at age 14 and switching to software :(

    Perhaps another question is whether it's a good idea to have the lead acid cell constantly trickle charging? At least with the solar panel the charging stops when the sun is down.

    I really appreciate any suggestions. I love these low voltage LED lamps and they are easily bright enough for what we need. Just the solar recharging is not working for us.
     
  2. wayneh

    Expert

    Sep 9, 2010
    12,087
    3,026
    I think you might consider using a LM317 adjustable regulator. You can bring the voltage down to what the existing circuit used to see from the solar panel. Using a meter, you could check that the resulting current flow was not excessive.

    It's hard to tell without the circuit you're using. I have solar lights that dump excess current once the SLA battery voltage is up to target. You may have something similar. It would be inelegant to be dumping a bunch of current all the time, so it would be worth spending a little time tweaking.
     
  3. gerty

    AAC Fanatic!

    Aug 30, 2007
    1,153
    304
    Would it be possible to remove solar cell from light fixture and remote mount it in the sunlight?
     
  4. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    Here is a circuit that will keep the battery at about 6.3V. It charges at 100mA.
    The zener is a 5W unit, which will keep it relatively cool while dissipating about 700mW.
    The LM317 will dissipate a little over 1W, so you should mount it on a heat sink.

    If the charging current (Ichg) is too high, change R1.

    R1=1.25V/Ichg

    COVER MY ASS CAVEAT:
    I don't guarantee that this will not damage your circuitry or your battery.:eek:
     
  5. #12

    Expert

    Nov 30, 2010
    16,248
    6,744
    @ post#3

    Yes, you can install longer wires to get the solar panel in better sunlight. At this low of a current, the voltage loss in the wires will PROBABLY be small enough to make no problem.

    Ohms per hundred foot: Voltage lost at 100 ma
    #18 wire = 6.386 .6386
    #16 wire = 4.018 .4018
    #14 wire = 2.524 .2524

    You must double the real world distance to account for needing 2 wires to carry the current. 25 feet up a tree would make the current flow through 50 feet of wire.

    edit: This site program deleted my spaces, so the columns aren't lined up correctly.
     
  6. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    Next time, try wrapping it in CODE:
    Code ( (Unknown Language)):
    1. Ohms per hundred foot:  Voltage lost at 100 ma
    2. #18 wire = 6.386            .6386
    3. #16 wire = 4.018            .4018
    4. #14 wire = 2.524            .2524
    5.  
    That's a good idea. better than mine.:p
     
    #12 likes this.
  7. johnmhome

    Thread Starter New Member

    Apr 19, 2012
    4
    0
    The solar panel does come with wires so it can be remotely mounted but there's still not enough sunlight. The house is in almost permanent shade. They're redwood trees, so really hard to get all the way up to catch the sun :)
     
  8. #12

    Expert

    Nov 30, 2010
    16,248
    6,744
    That pretty much whacks the long wire idea.
     
  9. johnmhome

    Thread Starter New Member

    Apr 19, 2012
    4
    0
    Can you explain what the Zener diode does in this circuit? I thought a zener is a simple voltage regulator, but we already have the LM317 doing that.

    Also what will happen when the light switches on and starts drawing current (it's about 8W) ?
     
  10. #12

    Expert

    Nov 30, 2010
    16,248
    6,744
    The 317 in this circuit is being used as a current limiter. The 6.8 volt Zener is the voltage regulator, then D6 reduces the 6.8 volts by about .6 volts to 6.2 volts.

    When the light switches on, it will use power from the battery, except the add-on circuit will provide about 1/10th of an amp so the battery doesn't have to supply ALL the current.
     
  11. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    The LM317 is being used as a 100mA current source, limiting the charging current. See the datasheet, Fig. 26.
    The zener limits the battery voltage. The charging current is diverted into the zener when the battery voltage reaches ≈6.3V.

    Obviously, 100mA cannot keep up with 8W when the lights are on. How long do the lights stay on after triggering?
    Do you know what the solar cell output current would be if you had normal sunlight?
     
    Last edited: May 3, 2012
    johnmhome likes this.
  12. wayneh

    Expert

    Sep 9, 2010
    12,087
    3,026
    Take a look at the current dumping circuit in this post. It's at the far right side of the schematic. The zener and resistors determine the spillover voltage, turn on the transistor, and then current dumps thru a resistor instead of the battery. This wouldn't work for a lot of excess power, well it would work but you'd just be wasting power in the resistor, but it's pretty slick to just trim the panel current a bit to protect the battery. This isn't my design - I studied the device and drew it.
     
    johnmhome and #12 like this.
  13. johnmhome

    Thread Starter New Member

    Apr 19, 2012
    4
    0
    The lights might be on for a couple of hours (one of them lights an outdoor dining area we use in the summer). The lead acid cell seems able to power this for long enough.

    I guess my biggest concern is the idea of permanently charging the cell. Any concerns?

    The solar panel puts out around 300ma, FWIW.

    Anyway, your suggestion here is excellent, and I didn't know you could use an LM317 as a current limiter. I'll proceed with this design and see what happens.

    Many thanks.
     
  14. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    As I mentioned in post #4, you can change the charging current. Charging will stop when the zener voltage is reached.
    You may need to select a zener to get the proper float voltage of 6.75V to 6.9V. I had chosen ≈6.3V, but that may be a little low. I based it on the voltage (12.6V) of a Harbor Freight maintenance charger that I used on my motorhome batteries.
    The series diode drop will be a few tenths of a volt when the zener is taking most of the charging current, so you have to pick a zener that is about 0.4 - 0.5V higher than the float voltage. A 7.5V zener might be better than 6.8, but I would probably choose to float a little low rather than a little high, if I had to choose. If you float it too high, the water will slowly boil away.

    EDIT: if you increase the charging current, you will have to scale C2 up proportionately, to keep the ripple voltage to a tolerable level.
     
    Last edited: May 13, 2012
Loading...