Electrical Gremlins

Discussion in 'Old's Cool' started by Malindi, Apr 19, 2004.

  1. Malindi

    Malindi Zen Adventurer

    Joined:
    Jul 9, 2001
    Oddometer:
    5,425
    Location:
    Vancouver, BC
    I know that trying to help someone diagnose an eletrical problem is as exiting as watching paint dry, but this one has me stumped.



    Bike : R80G/S, 1986

    Symptoms:

    2 weeks ago, intemittent headlight failure. Sometimes requires stopping, turing ignition half-way off and on to cure problem (bike is wired to allow headlight to be turned off with the ignition key, Euro sytle, also has K75 headlight bucket instead of stock).

    Same ride, problems worsens.

    Today, bike started fine as always, but no headlight, nada. Tried the old tricks (headlight switch jiggling etc.), all to no avail

    Removed headlight, checked socket and find 12V and ground when key is on. Bulb is fine. Plug in bulb, no voltage (no light) at the + for the socket. Suspect dead socket or bad connection. Remove socket tab, check voltage and find 12V. Connect bulb, no light, also no voltage on + for the socket when I turn things on!! Getting confused and try bulb on various other + contacts in headlight bucket. Works fine.

    Other observation: with the headlight removed, the highbeam light comes on when I “flash” or engage the highbeam with the toggle. With the bulb in I get no light, although on the “flash” setting the highbeam light comes on, but not with the regular highbeam engaged.



    Thing get worse. Halfway through working on this problem, I find that oil sender light, neutral and gen light are no longer working. Indicators, rear lights work fine.



    I removed the left handledbar indicator/switch and cleaned it out with th right stuff, removed the ball and spring and re-assembled in-situ (you haven’t lived until you try that, and succeed, when the indicator assembly is still attached to the bike).



    Fuses are fine.



    Either there’s a bad + connection, allowing 12V to pass with the bulb removed but not with the bulb in there, or there’s resistance to the extent that it passed a 12V reading but not enough juice to run the light. Or it’s a gandiose ground problem somewhere …



    Help appreaciated (I’m on postpone for the Airlist, so direct replies appreciated)



    K
    #1
  2. Humungus

    Humungus no it doesn't

    Joined:
    Jun 16, 2002
    Oddometer:
    5,313
    Location:
    circling just out of your campfire light
    Mate...thats a big call to sus this out. All i can suggest is the obvious...follow all wiring & look at the all connectors (open them & look at them) all bits of tape & very likely a wire being pinched somewhere. Redo any eaths to ensure good grounding.

    Sorry its a shot gun effect but electics can be fickle on a good day so you must ensure path is good.

    Good luck
    #2
  3. Jim Bud

    Jim Bud Long timer

    Joined:
    Apr 25, 2003
    Oddometer:
    3,439
    Location:
    Castle Rock, CO
    Different bike, similar problem.

    I suspect that there is an "almost open" connector somewhere.

    It is probably oxidized and allows 12 volts @ very low amperage to flow; hence, the meter reads OK, but wil not carry any current, hence the light does not work......

    I had a similar thing on another bike and I had to follow the circuit back and find a connector that was the problem...it can be a pain, because if similar, the connector will work sometimes ( when moved/disturbed) but then not work....you just have to trace it back with a wire connected to a light bulb so it draws some current and find that sucker..

    Good Luck
    #3
  4. Guzz

    Guzz Gutless wonder

    Joined:
    Feb 15, 2002
    Oddometer:
    4,934
    Location:
    Tempe AZ, USA
    Your on the right track. Start where the problem is, and keep working back. Get out your voltmeter and trace the wires to a connector, and check voltage (if you have any). If no voltage, keep tracing the wires back to the next connector.

    Keep on repeating the process till you find voltage. You found where the problem is. Repair accordingly (new connector, new wire, etc).

    It sucks, but be patient.
    #4
  5. Frank Warner

    Frank Warner Traveller

    Joined:
    Mar 14, 2002
    Oddometer:
    6,593
    Location:
    Sydney, Australia
    1) When you use the voltmeter .. take the negative lead to the bikes chasssie somewheres .. not to the bikes brown wires... that will mean your readings are true even if the brown wires have a problem.

    2) I'd suspect the ignition switch .. with th elight switched 'on' check for +12 volts on 'input' side.. if it ain't on the input then don't bother with the output side. If no +12 on the input side - fues contacts can get dirty .. check for +12 on the fues it self and then on the back of the fues terminals.. if that is OK then it is th ered wire - probably where it passes under the steering head.

    2a) Ignition switch if the +12 is ok on the input side but not on the output side it is the igniton switch .. take it apart and clean it. Put some electrical contact grease in there when finished to help keeep it going.

    3) if you get +12 everwhere then check the headlamp brown wire .. +12 volts here indicates the brown wire is broken .. it should be 0 volts ... broken either under the steering head or in one of the crimp/weld joints inside the wireing loom. Nasty.

    That might help.

    Yes Jim your on the right path .. intermitant resistive connection.
    #5
  6. Bart

    Bart Constant Lurker

    Joined:
    Sep 25, 2003
    Oddometer:
    2,963
    Location:
    Air Transport Central
    As an experienced electrician on Aircraft and bikes, Voltage is not always a good measure of Power. The reason being, if you have a wire that is corroded,pinched or broken except for a couple of strands the wire will read votage fine but will not allow enough current to put light on the situation.

    Do a physical check of the whole wire and connections. If there is no evidence of pinching you do not have to tear apart a wrapped bundle. Surely you will find the problem. The first thing I was taught was to look for physical causes and you will usually find the problem.

    I Hope this helps.

    Bart
    #6
  7. Malindi

    Malindi Zen Adventurer

    Joined:
    Jul 9, 2001
    Oddometer:
    5,425
    Location:
    Vancouver, BC
    Thanks to all who replied. It turned out to be dead positive lead from the ignition switch to the headlight switch in the handlebar. The break was between the headlight switch and the red connector under the tank. Said lead tested with 0 Ohm resistance, continuity, and conducted 12V, but once there was load, the 12V disappeared. Very odd … Never the less, I fed a new wire and all’s well now.



    I also noticed I managed to hit the kill switch in all my maneuverings the other day, which accounted for the idiot lights not coming on.



    K
    #7
  8. kevbo

    kevbo Rubbery-Lip Flappin' PHI

    Joined:
    May 29, 2002
    Oddometer:
    1,478
    Location:
    Albuturkey
    I see Kberretta found his problem, but perhaps having this in the archive will help someone else.

    What was found is typical of a high resistance connection or almost broken wire. Even if there is a fraction of an ohm (which can't be measured with a 2 wire ohmeter) the "choke point" may heat and open more fully under load.
    Also, poor connections are natures origional diode (used in crystal sets) so it may well read OK with the polarity of the voltmeter, but not pass any current to the load.

    The key is to trouble shoot with the load connected. Removing the bulb removed the load, and makes the problem impossible to detect.

    A voltmeter can be very useful to find this, but needs to be used in a different way. Use it to hunt down the voltage drop across the poor connection, not to "see where voltage is getting to"

    Example 1: Connect one lead to each end of a suspect wire.
    If the wire is good, you will measure near zero volts. If the wire is bad you will see the ~12V it is dropping.

    Example 2: Connect one lead to each side of a suspect connector. If the connection is good, you will see ~ zero volts.

    Now you don't have to check each and every possible component. Start by connecting leads to each end of the whole circuit. If you see the voltage drop, you know the problem is between the leads. Find some accessable point somwhere near the middle of the circuit, and try both halves. Now you know which half has the problem. With 3-4 such tests you can thus test 8 or 16 possible failure points.

    WARNING: This only works right if ONE thing has failed. On a basket case that set out in the rain for years, you might have 3 poor connections in a single circuit, and you will have to resort to brute force checking of each one.


    Finally a Big halliluja for all participants in this thread. NOBODY refered to the problem as a "short!" Which is a whole different kind of problem and requires different techniques to locate.
    #8
  9. Guzz

    Guzz Gutless wonder

    Joined:
    Feb 15, 2002
    Oddometer:
    4,934
    Location:
    Tempe AZ, USA
    :D Didn't you explain this once before, cause I remembered not to use the word "short" anymore. :D
    #9