Announcement

Collapse
No announcement yet.

Switching LED bulbs

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Switching LED bulbs

    I'm running 12 VDC LED bulbs in strings requiring 15 A @ 12 vDC. My question is about the best way to switch these strings. I've been using SPST toggle switches rated for 20 A @ 12 vDC, but have been experiencing some premature bulb failures. Should I be using a different type of switching for such DC currents? Am I introducing surge or arcing with simple switches?

  • #2
    When you say 15A string at 12V, are you running from a battery? Battery with a solar PV charger ?

    When LED bulbs are rated at 12V, applying 12.7V will shorten their life, and 13.8V will cook them in months
    Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister

    Comment


    • #3
      Originally posted by Mike90250 View Post
      When you say 15A string at 12V, are you running from a battery? Battery with a solar PV charger ?

      When LED bulbs are rated at 12V, applying 12.7V will shorten their life, and 13.8V will cook them in months
      Thanks for your comment Mike. Can you tell me more about the voltage impact upon LED lifetime? I have searched the web and spoken to LED suppliers and have not been able to get a quantitative answer on how high voltage can be above 12 vDC before longevity suffers, other than “higher voltage” can shorten LED longevity. Whatever than means. You’re the first person to put a number on it.

      I head up a volunteer group who install solar powered overhead LED lighting in remote centers for orphan kids in Malawi, Africa (6% of country has electricity). Typically an installation has 2 lighting circuits of 30A each, at 12 vDC, comprised of networks of 5 watt LED MR16 bulbs. The lighting circuits are the connected "load" from a Morningstar MPPT solar controller, and basically are wired in parallel with the battery bank. So, as you mentioned, while at night the system is at 12.0 vDC, during the day the batteries are seeing 13.8 to 14.4 vDC.

      With the general feeling that the higher voltages MAY be detrimental long term, we tell the people not to turn the lights on during the day. But who knows what happens when we leave? Not knowing how sensitive the voltage dependence might be, we did not before install any controls to prevent the lights from being turned on during daylight charging when the voltage was higher. We have 5 of these centers so far, but at our most recent installation in 2014, we just recently learned that 98% of the bulbs had burned out after 6 months of operation. Perhaps a bad batch of bulbs, or perhaps the center supervisor didn’t keep the lights off during the day, assuming those extra daylight volts are more significant than I thought. Hence the voltage dependence question.

      So if these voltages are affecting LED life, what’s the best way to control this? I leave for Africa in 3 weeks, so I’m under the gun to assemble and carry over a solution. On a different subject, I’ve been using manual SPST switches to turn the 30A-12 vDC circuits on/off, but could that be causing transient surges and affect the LED lamps?

      So I had decided that controlling the voltage the LEDs see is worth attending to, at least "directionally" Your comment adds strength to that argument. Here’s what I’ve begun to assemble: I could “lock-out” the lighting circuit switches with a photocell controlled relay, but decided a voltage sensing control and relay would be more reliable. I’ve purchased small inexpensive PLC controllers, and voltage transducers, but having a difficult time deciding what an appropriate 12 vDC relay would be to control 30A – 12vDC circuits. Probably better to break these circuits in half to 15A circuits, but still not sure what an appropriate relay would be for this service.

      Any help you could provide would be greatly appreciated, and I definitely wish to hear more about voltage impact upon 12 VDC LED lamps.

      Thanks

      Paul
      Attached Files

      Comment


      • #4
        Bare LED's are a simple diode. When fwd biased, they conduct and the amps must be limited, or the diode burns out.

        When a LED is fitted into a bulb, they can either use a DRIVER circuit to control the power, or they use a simple resistor. A simple resistor will allow more amps as the voltage goes higher. Then it's a race to the bottom, of how much heat sink is in the fixture vs the extra power dissipated by the diode junction. As heat increases lifetime decreases.

        What is happening in your case, depends on how the LED bulb is made. But simple DC wires and switches can't build a surge, coils or inductors need to be present for that to happen.
        Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
        || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
        || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

        solar: http://tinyurl.com/LMR-Solar
        gen: http://tinyurl.com/LMR-Lister

        Comment


        • #5
          Originally posted by Mike90250 View Post
          Bare LED's are a simple diode. When fwd biased, they conduct and the amps must be limited, or the diode burns out.

          When a LED is fitted into a bulb, they can either use a DRIVER circuit to control the power, or they use a simple resistor. A simple resistor will allow more amps as the voltage goes higher. Then it's a race to the bottom, of how much heat sink is in the fixture vs the extra power dissipated by the diode junction. As heat increases lifetime decreases.

          What is happening in your case, depends on how the LED bulb is made. But simple DC wires and switches can't build a surge, coils or inductors need to be present for that to happen.
          My bulbs have more than a resistor inside. I've had a couple break, and there's a pretty complex looking circuit inside.

          And OK, I get that . .. .Probably not a surge causing bulb failure. But let's return to your statement that an extra volt or two can burn out a bulb in months. That's a big point that I've never heard anywhere else. And may explain a lot of our problem, where voltages are applied to the charging batteries up to 14.4 vdc, if people refuse to listen and leave the lights on during the day. I'd like to know more about overvoltage.

          But what about under voltage? Can 11 vdc or less shorten LED bulb life?

          Comment


          • #6
            There are regulated LEDs that are suppose to be good from 8-30 volts. Expensive and put off RFI, I am glad i don't need them.

            Comment


            • #7
              Originally posted by jimindenver View Post
              There are regulated LEDs that are suppose to be good from 8-30 volts. Expensive and put off RFI, I am glad i don't need them.
              I am trying desperately to gather knowledge toward solving a very real problem of failure of existing "run-of-the-mill" 12 vdc MR16 LED bulbs. Can anyone provide me with some useful information on voltage vs longevity, or a reference I can go to?

              Comment


              • #8
                Originally posted by Paul_NJ View Post
                I am trying desperately to gather knowledge toward solving a very real problem of failure of existing "run-of-the-mill" 12 vdc MR16 LED bulbs. Can anyone provide me with some useful information on voltage vs longevity, or a reference I can go to?
                In general, incandescent bulb life is drastically reduced with excessive voltage. They sort of ride a
                critical point between where the voltage is too low to make much light, and where higher voltage
                burns them out fast. Bruce Roe

                Comment


                • #9
                  Originally posted by Paul_NJ View Post
                  I am trying desperately to gather knowledge toward solving a very real problem of failure of existing "run-of-the-mill" 12 vdc MR16 LED bulbs. Can anyone provide me with some useful information on voltage vs longevity, or a reference I can go to?
                  Does "run of the mill" mean off-brand ebay specials? Because if so, the premature failure might not be your fault at all. LED's aren't as simple as incandescents, and you'll need to know something more about how they are made to understand how they will react to different voltages. Are they dimmable? Are they intended for 12 Vac, and being used with DC? Really, what did you buy?

                  Philips has a white paper on LED longevity, maybe it will give you some more ideas.
                  CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                  Comment


                  • #10
                    Originally posted by Paul_NJ View Post
                    I am trying desperately to gather knowledge toward solving a very real problem of failure of existing "run-of-the-mill" 12 vdc MR16 LED bulbs. Can anyone provide me with some useful information on voltage vs longevity, or a reference I can go to?
                    There is no "run-of-the-mill" LED bulb. Every mfg builds them differently. I'll attach a link to MY favorite LED bulb, but it's not MR-16.
                    https://www.wiredco.com/10_LED_SMT_S...rmwhiteled.htm
                    These are replacements for reflector equipped bi-pin base with a 10-30V AC-DC rated driver built in.
                    So, something like this is what you you need, that is rated for a wide voltage range. You already know what happens with 12V rated LEDs on higher voltages.

                    Billy Bob's LED bulb factory does not make the LED chips inside the bulb. They buy them from Cree, Philips or another semiconductor wafer maker, and use the loose die as a basis for the LED bulb. The problem is to find what the die mfg rates their LED die for, and see if the bulb maker provides proper biasing AND heat sinking for the die. Over voltage or over heating reduces the life of the 20,000 hour die.
                    this mfg trade mag may help
                    http://www.ledsmagazine.com/led-desi...facturing.html
                    Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
                    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
                    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

                    solar: http://tinyurl.com/LMR-Solar
                    gen: http://tinyurl.com/LMR-Lister

                    Comment


                    • #11
                      So if these voltages are affecting LED life, what’s the best way to control this?
                      My thought would be a buck-boost DC/DC converters.

                      For example - http://www.powerstream.com/dc-buck-boost.htm

                      I don't think you'll find one that can do 15A or 30A.
                      The above says 8A - so you could do 2 of them @8A to drive half of what you'd run off 15A.
                      And there's some loss in using the buck-boost converter. So 1/6 to 1/4 of the power becomes heat instead of light....
                      As well as extra expense in equipment.

                      Also, if you're doing a DC/DC converter, another option that becomes available to you is to do 24V or 48V that drives the converter (which drives the 12V lights)
                      But then you will not have a backup where they can take the converter out and directly drive the lights if the converter dies.

                      Comment


                      • #12
                        Originally posted by foo1bar View Post
                        My thought would be a buck-boost DC/DC converters.

                        For example - http://www.powerstream.com/dc-buck-boost.htm

                        I don't think you'll find one that can do 15A or 30A.
                        The above says 8A - so you could do 2 of them @8A to drive half of what you'd run off 15A.
                        And there's some loss in using the buck-boost converter. So 1/6 to 1/4 of the power becomes heat instead of light....
                        As well as extra expense in equipment.

                        Also, if you're doing a DC/DC converter, another option that becomes available to you is to do 24V or 48V that drives the converter (which drives the 12V lights)
                        But then you will not have a backup where they can take the converter out and directly drive the lights if the converter dies.
                        Using 24V and a simple buck reg at the LED would be the cheapest and most efficient way. Bruce Roe

                        Comment


                        • #13
                          If you are using anything but documented, name brand LED assemblies, then you will ony be guessing at why you're having failures. Most "run of the mill" LED lights that flood eBay and such markets are built to be brighter and cheaper than the competitor and little consideration is given to longevity. They are very often poorly regulated, poorly heatsinked and poorly constructed.

                          The best thing you can do with LEDs like that is to run them at about 10-20% below the suggested current (voltage) and hope for the best. Heat kills LEDs and that is often the thing least considered when manufactures are determined to make them cheaper and brighter than their competitiors. At 1/2 to 1/3 the cost of a similar name-brand LED products, today's consumer will just go buy another cheap set when the first one fails. Sad but true.
                          Dave W. Gilbert AZ
                          6.63kW grid-tie owner

                          Comment


                          • #14
                            Thanks to all for your comments and insights. I'll provide some "collective" responses.

                            First, yes, I made the mistake of purchasing "discount" LEDs, considering cost first. Here's what we've been using, and the ones that failed:

                            TorchStar bulbs.jpg

                            Lesson learned : you get what you pay for.

                            Another thing I've learned is that we can do a better job managing temperature. We make our own light "fixtures", and mount them to the underside of roof trusses in large open gathering rooms (remember this is in Africa). We use 2" PVC pipe couplings as shrouds, to hold the MR16s in place and protect against errant soccer balls, etc. But we didn't provide ventilation. This week, in my shop here at home I inserted a thermocouple into the bulbs "fins" in a test fixture, and measured temperatures with shrouds, without shrouds, and with drilled holes, and seems like we can reduce bulb surface temperature by 25 degrees with ventilation holes. Not sure how much, but directionally that's got to help.

                            P1020036.JPG

                            temperature.jpg

                            We've found a Canadian LED supplier willing to work with us and provide us a good discount on bulbs, considering this humanitarian program. We will be using these bulbs as replacements this year.

                            https://initial-led.com/product/5-3w...-incandescent/

                            Finally, we've assembled a couple controllers that will lock out the on-off switches when the voltage get above 12.5 - 13.0 vdc during the daytime solar charging, to prevent overvoltage when people don't think to keep the LEDs off during the day.

                            We leave September 13 for Malawi. We don't know what caused the failures, but we're throwing the "kitchen sink" at the problem, all "directionally" good. Just the nature of trying to run 21st century technology in an 18th century world, 8000 miles from home, when you've got 3 weeks to do it, and 49 weeks till next time.

                            Thanks for the help. Any further thoughts are greatly appreciated!

                            Paul

                            Comment


                            • #15
                              Originally posted by Paul_NJ View Post
                              I'm running 12 VDC LED bulbs in strings requiring 15 A @ 12 vDC. My question is about the best way to switch these strings. I've been using SPST toggle switches rated for 20 A @ 12 vDC, but have been experiencing some premature bulb failures. Should I be using a different type of switching for such DC currents? Am I introducing surge or arcing with simple switches?
                              Your question reminded me of using diode protection when switching an inductive load with a semiconductor - like a relay coil. The diode protects the driver against a voltage spike induced due to the inductance.

                              However, your situation is not this. Large current spikes will damage LED bulbs, like if you switched a poorly regulated supply on suddenly. In the time it takes for the supply to regulate, there can be a transient current spike because the voltage is too high. That isn't really your case. You could experiment and see what the difference in current is for one of your bulbs at 14.5V vs. 12V.

                              I think your idea of not allowing the bulbs to switch on in the daytime during charging is a good start - just don't make it so they won't come on when needed. Good luck.
                              Last edited by Living Large; 09-06-2015, 09:50 PM. Reason: edit

                              Comment

                              Working...
                              X