Forum Discussion

tarnold's avatar
tarnold
Explorer
Nov 17, 2014

LED flicker

Just installed a LED strip light in bathroom and works fine off of 12v. Cranked up generator and light flickers. Only this new one, have about 10 other bulbs that work fine. This new strip is 2 speed, bright and brighter. Doesn't flicker as bad on low setting. Any simple ideas?
  • Unlikely that it will stop the flicker. Try the LED connected directly to the battery and see if it works there. The battery acts as a very large capacitor so there should be little ripple there.
  • I have one phillips T10 LED which shares some wiring back to the battery with other devices, one of them being this laptop which is powered by a DC to DC converter.

    The laptop causes this bulb, and only this bulb to flicker.

    The amp draw of this laptop really jumps around quickly when read by my GT power meter or my clamp on meter, and I figure it is doing so much faster than these tools can measure.

    If I turn the laptop off but leave it plugged in to charge the laptop battery, the flickering stops. Only when laptop is running does it flicker.

    I can live with it. I wonder how long the Phillips t10 bulb can.
  • LSCamper, the light works properly off of 12 volts. Problem with flicker is when on gen or shore power.
  • The Magnetek has two outputs. One goes directly to loads and is completely unfiltered with lots of AC ripple. The other side charges the battery, so that the battery itself acts as a large filter. I agree the best solution might be to retire the Magnetek, but you can also try moving the circuit that feeds the light over to the filtered side. I believe my old Magnetek had two fuses on the right that were on the filtered output.
  • "A capacitor is not a capacitor"

    Different microfarads of capacitance react to different frequencies, differently.

    A battery more or less is equal to two farads of capacitance. But a battery is indifferent to much of the developed ripple in chargers, alternators, etc. Even INVERTERS send ripple back through the battery feed wires. Use your oscilloscope and confirm this.

    A ten thousand microfarad capacitor is the right size for the right job in the right place when placed inline with a single output secondary (low voltage) line. It suppresses ripple from an alternator to the point where the "magnitude" of the waveform is insignificant. Sometime when you aren't busy connect an oscilloscope to your charger or to your power supply output. Then enable AC power to the charger. What do you see? Every damned transistor and integrated circuit that is not switched mechanically "off" sees the same thing.

    I'll harp about this again but in a different manner. Devices in an RV have a shorter lifespan than they do on AC power at home. I'm talking about devices with solid state components. Think about this. Why? It's because of inadvertent electronic noise, but more than noise - outright assault by transient voltages. It isn't just the converter that causes transients, it's the chassis alternator, AIR CONDITIONER MAGNETIC CLUTCH, the engine starter motor, and generator starter motor.

    If you have no heart issues, stick your finger on the starter motor side of a solenoid, crank the engine then release the key. After you pick yourself up off the floor, switch to the engine air conditioning magnetic clutch and repeat the exercise. These transients indeed "get to" refrigerator solid state components, 12 volt electronics and even inverter internals. The effect isn't instant, it's water dripping on the forehead torture. The greater the amplitude, the larger the area under curve of the ripple the more harmful the effects. I have measured 300+ volts DC from an air conditioner clutch. Use a scope and have your jaw drop to the floor when you see the effect of what a starter motor does to DC power.

    I cured Dometic and Norcold board failures, around 80% of them using transient voltage suppression. I clamped hot to neutral with a 180 volt bi-directional device. Where? At the AC heater element voltage supply.

    OEM does not address this issue. Why? Because OEM could care less. They do not attach oscilloscopes, they do not monitor entire systems and I can can well guarantee you they are not going to do a thing to extend the lifespan of their devices well beyond the point of warranty.

    I cut my teeth on this subject, curing problems with Raytheon and Furuno radar and color sonar electronics aboard commercial fishing vessels. Owners of million dollar vessels do not give a damn WHY their trips have to be aborted, they want the root cause stopped, cold and dead in it's tracks.

    The suppression technique using capacitors is not theory. It has been proven so many times it is a joke. When a vessel went from 10K electronic repairs to a hundred dollars over a several year period, and was repeated, vessel after a few dozen vessels, the chance of the capacitor technique being a fluke is reduced to near zero.

    Adding a capacitor to a system that already has had its electronic components tortured for years does not guarantee the tortured component will not fail an hour after the capacitor is connected.

    To me a 200 mv ripple level is ridiculous. I attached a cap to the new to me WFCO and I cannot detect ripple now.

    By the way, take some time and find out CONCORDE BATTERY'S philosophy on ripple. They do not mince words.

    Ripple belongs in a trout stream or on the shelf of a wino liquor store. Not in a battery system.
  • If there is enough ripple to pinchygiggle the LED strip there is enough ripple that makes it way to anything 12 volt solid state, and personally, that gives me the willies.

    Pinchygiggle:h
  • Well my 100 amp WfCo charging my bank
    There is no flicker in my LED strip lights I put in my fluorescent fixtures
    I would see it in the morning doing coffee and breakfast etc..
    Not there..nada..no flicker..I have never measured it for ripple maybe I will some day
  • It's s Left Coast thing.

    It means to screw around with something or somebody for no apparent coherent reason. Sarcasm directed at garbage design in this case. People -should-understand the performance ramifications of what they design. if they don't there's always an opening At Dunking Donuts or the local Lube Bay...
  • A few terms.

    http://en.wikipedia.org/wiki/Ripple_%28electrical%29

    Ripple is a periodic variation of DC at a the fundamental or multiple of the fundamental. In a full wave rectifier it is 120Hz. In a switcher it could be 10kHz to maybe 1MHz.

    http://www.tvss.net/trans/trans-x.htm

    Transient voltages are random and often caused be switching inductive loads.

    A capacitor of the correct value and type placed in the correct place will help to eliminate both ripple and transient voltages.

    LED flicker may be caused by a random fluctuating voltage, bad connection, bad LED, bad convertor, or other things that are slowly fluctuating. Ripple on a LED at 120Hz can not be seen with your eyes even though the light is varying.

    A slow fluctuating voltage may not be filtered out with a capacitor depending on the amount of current drawn and the length of time between fluctuations. The battery is a good filter at low frequencies, not so much at high frequencies.

    Putting the LED across the battery with the power on may tell you that the converter is bad ( a volt meter may help also). It may tell you that the LED flickers with the higher voltage at the battery from the converter. If it does not flicker then wiring to the filtered output of the converter may fix the problem. Anyway it is just another test to help narrow down the problem.