Load Equalizer Circuit for Led Turn Signals (bulb replacement)?

Discussion in 'General Electronics Chat' started by roundi, Jul 16, 2017.

  1. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
    Hi there,

    When replacing turn signal bulbs (12v 27w) with Led, you will experience the notorious hyper flash issue. Hyper flash is when the turn signal light flashes rapidly.
    The easy way to solve the hyper flash, is to install a load resistor (6 Ohm 50w resistor). This solution will make the resistor very hot, and can be dangerous in tight spaces like a motorcycle.
    Another way is to use an electronic circuit to create this load without heat , and I am looking for that , does someone have a lead on how to achieve this ?
    Thanks !
     
  2. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,921
    1,794
    You simply cannot create a load that doesn't dissipate energy as heat. One might possibly emit the energy as light but then you will have a very bright flash somewhere.

    What would be a better solution is to find a replacement flasher controller designed to run LED bulbs.

    Walmart has one, you may want to check your specific bike and flasher with vendors. I would be quite surprised if your local parts shop didn't have just what you need.
     
  3. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
    Unfortunately it is not available, the flasher of the bike is embedded with a security module and does not have a led version, I have already checked.

    This module does that: https://www.amazon.com/Custom-Dynamics-GEN2SSHD-Signal-Stabilizers/dp/B01MSAVD65 (No load, no heat, recycle 90% of power)
    I can confirm that it works as I have one and I wanted to know how it is done ? Any ideas. Thanks
     
  4. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    I have two guesses:
    1) The linked product seems to be only for Harley Davidsons. It may be designed to exploit specific characteristics of Harley wiring that none of us are aware of and cause the signal to oscillate properly with no/minimal load. In other words, it may not be creating a "load without heat" at all, but using some other trick to satisfy the circuit.

    2) Capacitors can create impedance, and therefore voltage drops, with minimal heat dissipation depending on the frequency of the AC they're passing. I can vaguely imagine a concept (but can't quite picture how it all goes together) where you would use some sort of high frequency oscillator (555, LC, whatever) to switch the voltage source on/off as a square wave, then pass that through a capacitor whose value is matched with the oscillation frequency in such a way as to pass the desired current, mimicking the current draw of the real load, but with nearly all of the voltage drop being across the impedance of a capacitor (one with low ESR for minimal heat dissipation.) I vaguely think that much makes sense, but then you still have to complete the circuit, and I'm pretty fuzzy on exactly how that winds up. If any of that made sense, hopefully one of the experts here can run with the idea. I'll probably try simming it in LTSpice in the near future to see if I can make sense of it, but no promises!
     
  5. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
    Attached is the wiring diagram of the rear harness, I have put a red circle where the Load Equalizer goes. Basically it is simple wiring that make the turn signal work, I don't think the signal has anything in particular.
    SecurityModuleWiring.png
     
  6. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    I was thinking of the internal workings of the signal and security module, not the wiring itself. According to your description, the turn signal output from that module flashes rapidly with too light of a load and more slowly with a heavier load. It's conceivable that there's some way, other than driving a heavy load, to "trick" that control module into blinking at the right rate. If that's the case, it would require knowledge of the circuitry inside the control module in order to know how to do it. The wiring layout of the bike wouldn't be the issue, it would be the inner workings of the control module.

    As for my other idea, it's probably too-good-to-be-true, like perpetual motion or free energy, but I'm gonna try to figure it out anyway. Even if it doesn't work, hopefully I'll learn something in the attempt. I'll let you know what, if anything, I figure out.
     
  7. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,541
    409
    My guess is that it uses the existing flasher signal as a trigger and then generates its own flasher pulses at the correct rate.

    The incoming pulsed signal starts a re-triggerable monostable which provides a steady output to enable an astable oscillator to flash the indicators. All this can be done with a couple of ubiquitous 555 timer chips.
     
    ebeowulf17 likes this.
  8. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    That sounds a lot smarter and more likely than what I was thinking.

    For an update on my capacitor impedance plan, I've hit my first obstacle. I had been thinking of simply using the square wave from a 555 to control a MOSFET switching the supply voltage through a capacitor to ground. I thought if I matched 555 frequency and capacitor value properly, I'd get a predictable impedance. It doesn't work at all as planned. I think the issue is square wave vs. sine wave, but I'm not sure. If I feed a sine wave at my chosen frequency through a cap with my chosen value, the impedance matches my calculations and I get roughly 2 amps through the cap. However, if I feed a square wave at the same frequency through the cap I get milliamps or less. I'm not sure if it's high frequency content of the square wave causing higher impedance, or some other issue, but it definitely rules out any "simple" versions of my idea working. I'll probably dig a little deeper for my own curiosity and to learn some more, but I'm not holding my breath for any major breakthroughs!
     
  9. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
  10. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    Good find! Based on their block diagrams, it looks like the mcu determines the flash rate (as opposed to an old school rc circuit, 555 timer, etc.) and that it employs a current sensing circuit on the turn signal outputs. I'm pretty sure that rules out any trickery specific to the control module. Looks like you really will need to provide a load similar to what the control module expects to see.

    If I'm reading that system correctly, the 555 idea mentioned earlier wouldn't work because, even if you get the signals you've replaced working properly off of the 555 circuit, the remaining lights (dash indicator, any lights still running directly from control module) would probably hyper flash in response to the low current draw on the new circuit. Mind you, I don't know that for sure - it's just an educated guess.
     
  11. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
  12. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
  13. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    I don't think you can run enough power through an LM741, or any op amp, to fool the Harley. If you did somehow accomplish that, the heat dissipation would be the same as with the resistor idea.

    Just how hot does the resistor get? If you put the resistor on a heat sink and spread the heat thinner, would it be acceptable?

    What about getting resistors around 4x the value (24 ohms instead of 6) and putting 4 of them in parallel? Each one would have 1/4 the heat to dissipate.
     
  14. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
    the temp is about 250 degrees C (482 degrees F)

    Space is very tight where the load equalizer should go

    Also I have found another Load Equalizer , and not for Harley , fits any vehicle with bulbs replaced by leds : http://www.vleds.com/vlr-6.html

    I think we could rule out that the module previously discussed above uses a particular signal from the HD Turn Signal Module
     
  15. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    Yes, I definitely agree on that point. I thought I said the same thing several posts ago in reply to the patent you found.

    Maybe I'll try to take another look at my oscillator + capacitor impedance idea. I got frustrated with it earlier when I discovered what an impact square wave vs sine wave input had on the circuit. My fear is that generating a sine wave to run through the capacitor will generate just as much heat as dropping the DC load across a resistor, but maybe not. I'll take another look at single-supply sine wave generator circuits and see if I can figure something out. I still think this idea is a long shot, but it's the only idea I've got.
     
  16. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    That one specifically says it's just a resistor in a better heat sink:
    No magic there, just a significant heat source!
     
  17. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
    Yes you are right I missed that one ... so back to square one :oops:
     
  18. davideather

    Member

    Dec 12, 2016
    33
    3
    what you can do is replace the flasher circuit with one intended for a truck - it is basically a relay with a capacitor and resistor as timing components. Ask at a garage or ask a motor mechanic
     
  19. ebeowulf17

    Well-Known Member

    Aug 12, 2014
    1,211
    179
    I believe in this case the flasher circuit is integrated with a security system and some safety systems, so that's not a great alternative unless you want to abandon those systems as well, or at least not let those systems control the lights anymore.
     
  20. roundi

    Thread Starter New Member

    Jul 16, 2017
    8
    0
    another vendor, writes this:
    "A “rapid-flash” is caused by the factory turn signal module not seeing the specific amperage draw that the factory installed 1156/1157 bulbs provided. Whether you have a Plug-n-Play module or the hard-wired version, Load Equalizer Modules all work the same. Now you understand the features of our Load Equalizer Module, let’s talk about how it works."

    This gave me the following idea:
    instead of wanting to put an "electronic load" , why not find a way to "trick" the module so it "thinks" that it is draw 2 Amps.
    An additional lead is that the module is also connected to the +12v of the battery (as shown in the attached pic) , so maybe it draws the needed current from the battery ....

    Badland Load Equalizer.png
     
Loading...