d'aw what have i done now (headlight suddenly dead)

[tahrey]

OK ... so ... headlight survived dropping and bashing into the back of a van, just being pointed slightly to the right. Finally got round to doing something about it.

Unscrewed, unplugged, all that, like changing bulb before. Set to with hammer to straighten the one adjuster mount best I could, then crossbar and large philips screwdriver bit to free and tweak the adjusters themselves.

Got it looking sort of straight again, went back, plugged it in, turned ignition on all good. Mounted it and put the main screws in just enough to secure it.

Started engine to go mooch up and down the road to check ... slight flicker ... rev the engine and it dims and flickers more... then cuts completely.

Argh, what?
Full beam and pilot light still work fine - just not dipped beam - and the bulb looks in good condition still. It's not exactly old anyway. Made sure only to handle it with clean cloth between skin and glass. All the connections look fine too. All I can think is either something's got disturbed as I put it back in (made sure to route all the cables to not get pushed, pinched or stretched though), or it's the handlebar switch which I also suspect of being something to do with the "dead" fuel guage bulb.

Trying to trace it and figure things out is going to be a devil of a job, and it's cold now. Plus it was late when I was finishing off so haven't had chance between then and now (despite it being a few days) to investigate.

Is there any well known, easily fixed CG hiccup this could be, or am I stuck playing Sherlock Holmes?

[truslack]

Put a new bulb in and see what happens. If you have a multimeter then test the voltage from the dip beam wire to earth. An old indicator bulb in holder would also work as a tester.
____________________
Current: Suzuki RG 125 Gamma, Honda H100, Triumph Tiger 800XC, Suzuki SV650 (minitwin)

[tahrey]

Right, been a couple weeks, in which time I have stripped the seat, side panels, dismantled the light and clocks, made sure everything was properly seated, undamaged and otherwise shipshape ... and got nowhere.

Swapping the bulb does dick-all. I'll try the multimeter thing but I think I can guess that the voltage between the "hot" contact and earth is going to be 12-14v when idling or near the limiter, and a stuttery lower voltage or even 0v when in the useful rev range.

It's the strangest thing. When everything was off, it worked fine. But when I got the last screw done up, the behaviour returned. So I think it's probably something in the lamp itself, but sometimes other parts (e.g. manipulating the rear relays or handlebar switch) have an effect, and nothing I do to the in-lamp wiring can provoke the failure outside of its usual appearance. Only when it's largely or fully together, the engine is revved to what I guess is about 4000-6000rpm, and I'm least expecting it/feeling happiest that it's fixed, does it blip out. Something is unhappy about being vibrated at 75~100Hz or whatever its housing is resonating at... but, what?

Though I have my multimeter and I've done some soldering before, I'm largely an electrical testing newb. Is there a simple enough way of testing it all without stripping the whole loom out, or should I just throw it at a local garage and stump up whatever the labour charge is? I reckon I've easily lost that much already thanks to having to drive for work rather than ride, after all.
(If it was summer, no problem, but... yknow)

brainwave: if it comes down to testing the continuity of the headlamp flylead itself, I'm going to have to commit seppuku or something.

[N cee thirty]

swap round the wires that go to the headlight

it does all sorts if not plugged in right
____________________
'00 Aprilia RS50 > '92 Honda CG > '99 Yamaha Fazer > '91 Yamaha RXS > '79 Suzuki X5 > 01' Honda Cg > 07' Honda Cg > 82' Kawasaki Z200 > suzuki gsxr 400 gk73a > honda vfr 400 NC30 Mod 2 Passed 09/06/2011

Jewlio Iglesias wrote: I actually did vote BNP once

[tahrey]

Well it was definitely giving a dipped beam and a full beam as-is... i had a suspicion they were back to front originally because the bulb that i bought it with was worn out and all melty inside.

Still, I think I may have figured it. If my dodgy diagnosis is correct, it's an irritatingly simple thing. To whit, the halogen bulb I put in has contacts that don't mate up with the ones in the bulb holder very well - and it's either a design/manufacture issue, or something that's come about because of the bump, as two of the same type show this behaviour ... but when I put my remaining spare vacuum-incandescent in, it works fine.

(It was a game to get them to even fit properly in the holder in the first place, after all... maybe some extra work needs to be done, maybe sanding down the contacts to a different, blunter shape or whatever? Or adding a layer of softer solder on top?)

Took my multimeter and pulled it again. Had quite a bit of trouble getting readings because a/ the battery's running out , b/ the electrodes seem a bit too chunky to fit in the back of the plugs, so I had to go in the other way, and it was right fiddly. In the end, best as I can tell, the bike's feeding a steady 14-ish volts to the connector (didn't have a third hand, or assistant, to rev up whilst I did it though), but something wierd was going on with the headlamp and flylead. There was continuity - or at least, measurably low resistance - between pretty much all of the pins in the plug with bulbs in place (makes a certain amount of sense if there's no overriding polarity) EXCEPT for one - the dipped beam positive feed. It had continuity with NOTHING. Which makes NO sense, as it wouldn't work at all if that was true.

Checked both bulbs, which read as expected (3-4 ohm for the pilot light, 1-2 for the main bulb... though it still showed a path between the dip and full contacts?) and the holder connections vs the plug (dip-dip main-main earth-earth). No issues. Arrgh. The contacts however look rather vicious and primitive (just solder-covered pins sticking straight up) and the solder pads on the bulb slightly chewed up but not as much as I remember the old one being.

So... swap it out... rev it for quite a while, including doing a bit of 1mph speedway on the iced-up driveway... not a flicker.

So yeah. Stupid if that's true. I'll maybe brave the sidestreets to find a gritted but unlit main road when it gets a bit darker and see if it keeps working and has a usable amount of brightness/reasonably tight pattern.