Five blown lights -- one night; regulator stuffed?

[xwhatsit]

Left work, coming home about an hour ago. As I was putting my gloves and helmet on while the bike was warming up, I noticed I had no tail-light. `Oh boy here we go again'. However there was no brake light either. Went to nearest gas station, then another one, bought a bulb and replaced the burnt-out one. I thought maybe it just came to the end of its life.

Next one to go was my headlight, which I didn't notice for a while through the street-lit roads, and if it seemed a bit dim I put my hand in front of the headlight and saw light reflected, so it seemed OK. The light reflected was only the parking light. High beam still works and that's what I used to get home.

Then I put on my left indicator, and nothing happened at either the front or rear of the bike.

Obviously all these bulbs didn't just wear out on their own within the space of 30-45 minutes of riding. I have three thoughts:

When I did my re-wiring to make the tail-light work a week or two ago, something is shorting. This doesn't seem right, though; I don't think that would cause bulbs to blow anyway, and the fuses are all sweet
The bulb I replaced the blown one with is somehow causing problems. It is 21W/5W compared to the old blown one which is 23W/8W. This doesn't seem right either.
My regulator is blown. Will test what voltage I'm getting over the battery terminals tomorrow.I really hope it's not #3, but I have a spare (I think). However, has the regulator just gone as it just `wore out'? Or, are all the people who advised me not to replace my battery with a capacitor going to say `I told you so'?

Thanks for your help.
<hints id="hah_hints"></hints>

[xwhatsit]

Thinking again now, it's possible that the headlight blew earlier in the day (when I rode to work).

So if I was riding around all day with headlight switched on but only the wee little parking bulb drawing current, will that cause other bulbs to blow? No of course not, I'm talking shit.

Other ideas; bad connection to capacitor (where you normally have a battery), so the lights are getting flickery voltage, causing them to blow (like when you flick the light switch on and off at home until either your dad smacks you around the head or the light blows).

EDIT: Found my spare regulator (and, surprise! another ignition barrel, which is great -- and my bastard key doesn't work in it, does it?). It's a big dusty thing called `Shindengen SH232 12 0.7'. Two lots of three wires. Heaps of very similar looking things on Trademe. I know you can mix and match car regs -- can you do the same with motorcycle ones? Just in case I blow up my spare or it turns out to be no good.

SUPER EDIT: Apparently CX500 reg/recs never die, so I eBayed for them. Turns out they're the same model as the CB250RS reg/recs. Bollocks. Oh well, there's vast quantities of them floating around so I can get one for cheap if the spare blows up.

[bimotabob]

Hi

You are right "shorts" don't affect bulbs and the different tail light spec won't
cause it.
If it was me I would rip out that condenser and throw it as far as strength allows and buy a battery! this ain't rocket science.
Then I would do a charging check, if the voltage was over or under the specified amount (13.5-14.5 volts) I would check the stator is not earthed and do a open circuit voltage test on each phase (I think it's 3 phase??)
Then replace the reg and recheck.
Most old school 3 phase regs work the same, bridge rectification then zenor diode for each phase to stop voltage over 14.7.

3 phase in. often yellow
1 to battery positive often red
1 earth. often black or black/white tracer maybe green.
1 ignition positive in. Not always used but easy to wire in.

Blowing bulbs is annoying and dangerous but i'm sure you will sort this out if you haven't already.

Cheers
BB

[xwhatsit]

I thought the black wire on the 6-wire regs was for `voltage sensing', i.e. check the voltage running through the system? Yep it's 3-phase.

Going to go and check out the reg/rec as soon as I've had something to eat. If it's stuffed I'll just see the voltage increasing with RPM.

Les said something interesting, and that's that the headlight bulb might've caused the others to blow. Apparently regulator circuits don't react very quickly, and don't cut in til about 16V, so just by having the headlight go (and then leaving the switch on for ages) then the generator is pumping everything straight into the smaller bulbs. I know the tail-light and speedo light looked flaming bright, despite the tail-light being smaller voltage.

Oh well I'll check it out in a bit. Not really the most robust design in the world, motorcycle electrics.

[xwhatsit]

Well, fingers crossed, it's sorted.

Started up the bike, with headlight on low-beam (which has blown). No left indicators like before, try right indicators and they both go flash-bang. Tail light is working but is extremely bright. Open up indicators and all four indicators have their filaments plastered all over the inside of the globes.

Put my spare headlight bulb in, works great, tail light seems normal now, so do speedo backlights. Checked battery voltage; never exceeds 15V, even at 7000rpm or so. However it does seem very flickery, dancing all over the place. The digital multimeter can't quite keep up with it, so it's hard to tell (this is where I wish I had an analogue one), but it seems to go like 13V down to 8V then back up to 14V etc. At any rate it's not getting any higher than 15V or 16V or so, so the regulator seems to be doing its job.

Rode to local gas station, put in new indicators, seems to be absolutely fine. Here's hoping it stays that way.

So I think the headlight blowing seems to have caused the others to go. I think the capacitor is partly to blame for that; a battery would provide a much better voltage sink and I think the headlight going would have less effect. I think it'd be wise to carry a spare headlight bulb with me from now on.

The flickery voltage is probably to do with the use of the capacitor. I don't doubt it'd shorten bulb life as well, I think somebody brought that up in the capacitor thread. The 'lectrics will get a test tonight when I go to work, so I hope they'll hang together.
<hints id="hah_hints"></hints>

[bimotabob]

The wire colours bit was from my Suzuki experienace - your Honda may well have different colours, test to confirm.
So are you ordering the new battery today then before more bulbs are required?

Regards

[xwhatsit]

Hell no! :D

Regulator output is fine. Voltage is flickery, which almost certainly will reduce headlight life somewhat, but for all I know that bulb has been in there for years and I've done at least 2000-3000kms since putting that capacitor in, with headlight on 100% of the time.

I'm 99% sure that as Les suggested to me, that the headlight blowing took out a large voltage sink, so the other bulbs got hit directly with all the voltage from the alternator. Quite common on old bikes before reg/recs, and similar behavior is not surprising with the primitive reg/recs on bikes of my age without a large battery to soak up extra voltage.

I'm carrying a spare headlight bulb with me now, and making sure the capacitor is fully drained before I turn the ignition off.

Putting a battery in would be wise, but that would be giving in :lol:

[Coldrider]

It is over voltage (and vibration) that kills bulb filaments.
Generators & alternators do not pump current into bulbs & devices, the current is drawn by the resistance (ohms) of the device which determines the watts.

[xwhatsit]

It is over voltage (and vibration) that kills bulb filaments.
Generators & alternators do not pump current into bulbs & devices, the current is drawn by the resistance (ohms) of the device which determines the watts.

Probably vibration in my case, although a fluctuating voltage surely can't be good either. Standard headlight is very poorly mounted, basically the bracket is an interference fit to the bottom of the triple tree in rubber; the rubber perishes and it's no longer a proper fit.

I'm aware devices only draw the current they need, but the way it was explained to me the headlight draws voltage, so the overall voltage level in the lights circuit is raised killing the bulbs by over-voltage as you said. <hints id="hah_hints"></hints>

[Coldrider]

Yes, the headlight is being used as a shunt to load the alternator, normally the battery would do that.
Filaments also do breakdown over time, and excess vibration when the filament is white hot will eventially pop it.
I guess time will tell if the issue is resolved.

[smudge]

Yikes, hehe. If the voltage is all over the place as you describe, your battery wont last long and you may even end up boiling it and spilling acid ....
If you have replaced the battery with another device then you would need the knowledge to design the charging system and you wouldn't have the need to fault find it on here, I'm not trying to be mean, that's just how it is.

It is possible to have total loss type systems without running a battery but then as soon as one of the lamps blows you will get a higher voltage supplied to the other lamps and of course when your lights are off any other electrics will see high voltages. I used to take the battery out of my enduro bikes fit a wire wound resistor and put my own wiring loom in - dont ask why, no point really but it was simple and it worked.

[xwhatsit]

Yikes, hehe. If the voltage is all over the place as you describe, your battery wont last long and you may even end up boiling it and spilling acid ....
If you have replaced the battery with another device then you would need the knowledge to design the charging system and you wouldn't have the need to fault find it on here, I'm not trying to be mean, that's just how it is.

It is possible to have total loss type systems without running a battery but then as soon as one of the lamps blows you will get a higher voltage supplied to the other lamps and of course when your lights are off any other electrics will see high voltages. I used to take the battery out of my enduro bikes fit a wire wound resistor and put my own wiring loom in - dont ask why, no point really but it was simple and it worked.

<hints id="hah_hints"></hints>People don't have to answer my questions here about electrical fault finding. If they think I should work it out by myself because I made the modification myself then they're free to ignore my threads.

As far as I know the capacitor replacement of a battery was a fairly common thing to do, especially with magneto ignitions. It was all discussed in-depth in the original thread. Of course problems might come with that modification, but I am by no means demanding help. Forums are about sharing information, and many people are only too willing to add to the KB pool of knowledge by responding to my questions here.

If you think I should just keep quiet and try and figure it out on my own because it's a modification that I did and therefore I should know everything about the subject, then don't respond to my questions. I don't see that as `just how it is', I think that's completely contrary to the spirit I've seen from most motorcyclists.

[jonbuoy]

If it was an old surplus store cap it might have dried out (lost its juice and capacity) and be up to F*CK all mate. If you have a digi multimeter see how many AC volts as well as DC volts you have across the cap.

[xwhatsit]

If it was an old surplus store cap it might have dried out (lost its juice and capacity) and be up to F*CK all mate. If you have a digi multimeter see how many AC volts as well as DC volts you have across the cap.

<hints id="hah_hints"></hints>The cap is fine, I think -- I get a good 5-6 seconds or so out of dash lights and ignition system on, much less of course with the headlight on. I'm thinking the flickery voltage is due to delayed reactions from the voltage regulator; once the cap hits full-charge, the reg cuts the power supply. The cap drains (relatively) quickly, so the reg switches back on and charges the cap quickly again. With a battery, the charging would happen much more slowly, and so would the draining, so you'd get a much lower-frequency and lower-amplitude voltage flicker, you probably wouldn't notice it so much.

Bulbs have lasted ~500kms since I replaced the headlight bulb, so I'm picking the problem is solved for now. I'm happy to carry a spare headlight bulb around with me, so this current situation is more than liveable. Over the break I may try and design a real voltage regulator in the proper sense of the word (instead of the curious automotive variety), that imposes a hard limit on the amount of volts. Depending on the amount of current I'll have to pass through it, I may be able to get by with a cheapie off-the-shelf SMPS IC.

[howdamnhard]

Scrap the cap and refit the battery.:yes:

[davereid]

Your alternator produces AC, which is passed through a diode pack (Rectifier) to convert it to a series of half-wave pulses.

Battery charging.
As all these pulses are positive going when compared to the ground of the frame, they are suitable for charging a battery.

DC power.
This "rectified AC" as it is called is not really DC though, as it goes from a value of zero, to its peak value (about 19 volts) and back to zero again, many times a second. (The exact rate depends on your engine rpm.)

The battery and capacitor.
If you have a battery installed, the battery will tend to charge during the period where the rectified ac wave is greater in voltage than the battery. The battery will discharge, but maintain system voltage when the rectified ac wave is at a lower value than the battery voltage. But the battery resists going much above its nominal voltage, and helps stabilise the system voltage, normally at about 70% of the peak voltage.

A capacitor on the other hand tends to charge to the peak of the applied rectified ac waveform. So bikes running a battery free system tend to run a bit high in voltage, and bulbs particularly become sensitive to vibration and like to fail at switch on, or at high rpm.

Your regulator, and the average life of your bulbs will improve if you can find a bit of extra load to add to the bike.

An extra 12-20 watts would help the problem go away. This is where the idea of the wire wound resistor comes from. Even increasing the wattage of the bulb you use for your headlight would help.

[smudge]

Didn't mean to upset you, apologies if I did. I don't know the history behind it all, just trying to help a fellow motorcyclist...:scooter:

One thing I can tell you is that with these type of systems if it isn't all 'balanced' it will suffer from over voltage problems, as I said before I'm a sparky who has done this stuff to bikes before. You are getting some good advice here from the other guys so I'll butt out.

[xwhatsit]

Didn't mean to upset you, apologies if I did. I don't know the history behind it all, just trying to help a fellow motorcyclist...:scooter:

One thing I can tell you is that with these type of systems if it isn't all 'balanced' it will suffer from over voltage problems, as I said before I'm a sparky who has done this stuff to bikes before. You are getting some good advice here from the other guys so I'll butt out.

Sorry mate, I misinterpreted what you were saying and over-reacted. I need to drink more coffee :yes: Oh well, at least I used up my weekly mindless rant.

@davereid: Cheers for that, very enlightening. I realised the DC was really a cut sine-wave (^^^^^^^), but I didn't understand the relevance for this when comparing batteries and capacitors. I'm sure I can find something that will sink 20W in my box of bits. I'm still leaning towards a `real voltage regulator' as it seems a more elegant solution, but that will probably come with its own set of difficulties.

Thanks guys.<hints id="hah_hints"></hints>