Lighting coil issue, help me please

[Starky307]

I've just tried fitting a lighting coil (alternator) to my KXF250.

The output from the alternator goes to a rectifier, it then goes through a diode to a battery, just before the diode it feeds a coil (for the headlight contact). The battery (down stream of the diode) goes through the contact (of the coil) and then to the headlight and back to a common negative on the coil.

It won't bring the coil in but it will charge the battery, there appears to be no potential across the coil (which is the dc output of the rectifier before the diode supplying the battery).

When the bike runs the voltage across the battery increases so the coil output is working.

The idea is for the lights to only work when the bike is running without the need for a switch.

Any ideas what wrong?

[CookMySock]

Hokay.

Is this a single phase or three phase stator ? Can you post a wiring diagram?

What do you mean "coil" ? Do you mean a coil of a relay? Do you have a test meter or test lamp? Take some measurements across the relay coil and tell us what you see.

Try placing a small bulb in line with the relay and see if it lights up a wee bit.


Steve

[Starky307]

Hokay.

Is this a single phase or three phase stator ? Can you post a wiring diagram?

What do you mean "coil" ? Do you mean a coil of a relay? Do you have a test meter or test lamp? Take some measurements across the relay coil and tell us what you see.

Try placing a small bulb in line with the relay and see if it lights up a wee bit.


Steve

Single phase, unable to sketch or scan a wiring diagram on my apple mac book. I mean a relay coil, yes to meter and lamp.
Across the relay coil is the resistance of the coil, I'll need to measure it tonight as I can't remember what it was.

I'll try putting the test lamp in series with the relay coil tonight and see what happens, I'm not thinking much will happen as the voltage measured between those points last night appears to only be the difference between the battery voltage and the rectifier output. By this I mean at idle there is 0 volts and at high revs it gets to about 1.5 - 2 Volts dc, so i would guess with a lamp in series it would not go and then at high revs it would be very dim.

I'll try and sort out a wiring diagram of what I have done.

[Starky307]

Picture added, please excuse the quality as I tried to draw using a free ware version of photoshop. I think you'll get the idea.

[CookMySock]

Ok you are sort-of on the right track. I just wonder if the layout is slightly wrong. Will need to see a sketch I think.

Steve

[paturoa]

Can you label your diagram with letters and then tell us what the voltages are using those references.

Also set your meter to AC volts and do the same.

Are you sure that it is just a rectifier? Could be a regulator, still have to wonder why the diode is in circuit unless the intent is to use as a switch to stop the light when the engine is off.

If you are absolutely sure that its just a rectifier, also test the voltages without the battery connected, but don't rev it too much or you might blow the bulb.

Edit - can you also draw if there is a switch for the relay. Edit x2 - re-read first post Duh!

[bogan]

diagram layout looks good as to me. Those voltage figures suggest you may have got some wires hooked up a bit wrong perhaps? sure +ve of coil is to rectifier positive output, and -ve of coil is to ground?

[bogan]

Can you label your diagram with letters and then tell us what the voltages are using those references.

Also set your meter to AC volts and do the same.

Are you sure that it is just a rectifier? Could be a regulator, still have to wonder why the diode is in circuit.

If you are absolutely sure that its just a rectifier, also test the voltages without the battery connected.

Edit - can you also draw where the switch is for the relay.

if i understand correctly, the diode is the switch for the headlight coil, ie it stops the battery from operating the headlight coil, and when running power from the alternator/rectifier activates the lighting coil, a smart and simple idea :niceone:

[CookMySock]

if i understand correctly, the diode is the switch for the headlight coil, ie it stops the battery from operating the headlight coil, and when running power from the alternator/rectifier activates the lighting coil, a smart and simple idea :niceone:On the face of it, thats a good idea, but it will drop 0.6V from the battery state, quite likely removing as much as 30% charge from the battery.

I'm surprised that doesn't work though. I would test the relay to make sure its ok. Take a voltage measurement right across the relay. Also, perhaps you have the relay wired incorrectly.

If none o that works, try removing the diode and wiring the output of the rectifier directly to the battery, and then connecting the relay coil straight across the AC output of the stator.

edit: you might be able to use the diode to rectify the stator directly to the relay coil, but the relay coil should work from AC I would have thought. The half wave rectification the diode will provide will not provide much power to the relay coil so try with and without it.

Steve

[bogan]

On the face of it, thats a good idea, but it will drop 0.6V from the battery state, quite likely removing as much as 30% charge from the battery.

good point, unless he uses a schottky diode then itll only be .15V

[Starky307]

Can you label your diagram with letters and then tell us what the voltages are using those references.

Also set your meter to AC volts and do the same.

Are you sure that it is just a rectifier? Could be a regulator, still have to wonder why the diode is in circuit unless the intent is to use as a switch to stop the light when the engine is off.

If you are absolutely sure that its just a rectifier, also test the voltages without the battery connected, but don't rev it too much or you might blow the bulb.

Edit - can you also draw if there is a switch for the relay. Edit x2 - re-read first post Duh!

It's a regulator rectifier all together.

I'll try and do the drawing with reference voltages.


if i understand correctly, the diode is the switch for the headlight coil, ie it stops the battery from operating the headlight coil, and when running power from the alternator/rectifier activates the lighting coil, a smart and simple idea :niceone:

That is exactly what I wanted to happen, just like the klx 450 does, no light switch, light only works when the engine runs.


On the face of it, thats a good idea, but it will drop 0.6V from the battery state, quite likely removing as much as 30% charge from the battery.

I'm surprised that doesn't work though. I would test the relay to make sure its ok. Take a voltage measurement right across the relay. Also, perhaps you have the relay wired incorrectly.

If none o that works, try removing the diode and wiring the output of the rectifier directly to the battery, and then connecting the relay coil straight across the AC output of the stator.

edit: you might be able to use the diode to rectify the stator directly to the relay coil, but the relay coil should work from AC I would have thought. The half wave rectification the diode will provide will not provide much power to the relay coil so try with and without it.

Steve

Battery voltage increases to between 13.5 and 14 volts dc with half or more throtle, drops back to 12.5 when no engine running.

I get no voltage (dc) measurement across the relay (I would expect the same voltage here as the battery when engine running) but then it will sometimes be the voltage difference between the battery without the engine running and the battery with the engine running.

Relay works, as the system works on the bench or if I short out the diode (back feeding from the battery).

The ac output is unclipped prior to the regulator/rectifier so I don't want to put it to the relay coil.


good point, unless he uses a schottky diode then itll only be .15V

I've tried two diodes and both the same result, they are fine as I have measured them and they are around the 0.567 v breakover, what I would expect them to be.


Keep the ideas coming

[Starky307]

All these measurement are from memory and with the bike running.
First letter is the positive lead

B-A 0 to 1.5 volts dc (depending on throttle position)
C-A 12.5 to 13.5 volt dc (depending on throttle position)
A=D
E-F 13-14.5 volts AC RMS
H-A = C-A

That is as much as I can remember.

I'm off to have another go at it so i'll try and bring back some more exact readings

[bogan]

one thing is maybe the stator/rectifier cant supply enough current to turn the relay on, though this seems very unlikely as it can supply enough current to power the headlight, it would only be if the voltage coming from the rectifier is unregulated, ie just a rectified half ac waveform. In this case the coils inductance woul provide significant impedance and there would not be enough current to switch the relay. A solution to this would be to put a large capacitor between A and B.

A quick test to check whether it is the power supply would be too clip a 12V source (other battery, or battery charger) to A and B, if relay clicks and lights come on then you have a power supply issue, may be able to fix it with a cap as above.

[Starky307]

one thing is maybe the stator/rectifier cant supply enough current to turn the relay on, though this seems very unlikely as it can supply enough current to power the headlight, it would only be if the voltage coming from the rectifier is unregulated, ie just a rectified half ac waveform. In this case the coils inductance woul provide significant impedance and there would not be enough current to switch the relay. A solution to this would be to put a large capacitor between A and B.

A quick test to check whether it is the power supply would be too clip a 12V source (other battery, or battery charger) to A and B, if relay clicks and lights come on then you have a power supply issue, may be able to fix it with a cap as above.

If I put another 12v source on A and B the lights come on.

I think the capacitor trick is what I am missing to bring the relay coil on.
Back to the shed to see what I can dig up.

I remember that instead of a battery I could of used a capacitor with the DC system instead of the battery, I just didn't know what capacitance to use or where to get one from bigger enough.

All this so I could use the trick LED tail light, should have just run the normal regulated output and had an AC headlight...:doh:

[CookMySock]

B-A 0 to 1.5 volts dc (depending on throttle position)
C-A 12.5 to 13.5 volt dc (depending on throttle position)That description is infeasible. You have an error somewhere.


E-F 13-14.5 volts AC RMS
H-A = C-A

That is as much as I can remember.

I'm off to have another go at it so i'll try and bring back some more exact readingsUh huh. :yes:

I think either the diagram you have drawn is not what you actually have, or else you have a fault in the circuit. Give all the connections the yank test and see if they fail.

You are on the right track though. What you have drawn should work.

Steve

[bogan]

wait a sec, this is not an electric start batteryas i first thort is it? and is this a complete different wiring set from the ignition stuff?

[Starky307]

wait a sec, this is not an electric start batteryas i first thort is it? and is this a complete different wiring set from the ignition stuff?

Complete system I have built from bike specific parts.

My bike is a MX bike that I brought a stator and reg/rec for.

It had no wiring diagram for the reg/rec so I made my own wanting to have an automatic headlight/tailight.

[Starky307]

That description is infeasible. You have an error somewhere.

Uh huh. :yes:

I think either the diagram you have drawn is not what you actually have, or else you have a fault in the circuit. Give all the connections the yank test and see if they fail.

You are on the right track though. What you have drawn should work.

Steve

B-A represents the voltage that is up and above that of the battery when the bike runs. If the bike is idling the bat = 12.5 and B to A = 0 when the bike is half rev or greater the battery voltage is 13.2 (+/- 0.3) and B to A = 1.3 or slightly more.

I have tried and tested all connections again.

The wiring diagram i have provided is exactly as it is wired on the bike, me and my Father have checked it again and again.

[CookMySock]

B-A represents the voltage that is up and above that of the battery when the bike runs. If the bike is idling the bat = 12.5 and B to A = 0 when the bike is half rev or greater the battery voltage is 13.2 (+/- 0.3) and B to A = 1.3 or slightly more.Impossible! The only way that can happen, is if the battery, diode, and relay coil is wired in series. The measurements you show, and the way the circuit is behaving, all lead to an error in the relay wiring. Your diagram shows it wired in parallel, which is correct. You have something wired incorrectly.


The wiring diagram i have provided is exactly as it is wired on the bike, me and my Father have checked it again and again.

If it were me, I would pull it all off and start again. Wire up the rectifier/regulator and the diode to the battery, and test that its working. Leave the headlight circuit completely disconnected for now. Add the relay coil and don't do anything else until this is working.

With the diode there, does it not occur to you that you can simply attach the wire to the headlight between the rectifier/regulator and the diode? No relay is required.

Steve

[bogan]

Impossible! The only way that can happen, is if the battery, diode, and relay coil is wired in series. The measurements you show, and the way the circuit is behaving, all lead to an error in the relay wiring. Your diagram shows it wired in parallel, which is correct. You have something wired incorrectly.

I think hes got a half ac waveform coming out of the rectifier, which is probably doing funny things with the relays coil inductance.

Why dont you just get rid of the battery, add a cap across the rectifier and connect the headlight to that? no need for relays etc, the headlight will only go when ur riding.

[CookMySock]

I think hes got a half ac waveform coming out of the rectifier, which is probably doing funny things with the relays coil inductance.I assumed a full-wave rectifier. Yeah, maybe.


Why dont you just get rid of the battery, add a cap across the rectifier and connect the headlight to that? no need for relays etc, the headlight will only go when ur riding.What he has drawn should be working though.

Steve

[Starky307]

I think hes got a half ac waveform coming out of the rectifier, which is probably doing funny things with the relays coil inductance.

Why dont you just get rid of the battery, add a cap across the rectifier and connect the headlight to that? no need for relays etc, the headlight will only go when ur riding.

I would love to do this but I don't know what size capacitor to use, any ideas how to size it up?

[Starky307]

Impossible! The only way that can happen, is if the battery, diode, and relay coil is wired in series. The measurements you show, and the way the circuit is behaving, all lead to an error in the relay wiring. Your diagram shows it wired in parallel, which is correct. You have something wired incorrectly.



If it were me, I would pull it all off and start again. Wire up the rectifier/regulator and the diode to the battery, and test that its working. Leave the headlight circuit completely disconnected for now. Add the relay coil and don't do anything else until this is working.

With the diode there, does it not occur to you that you can simply attach the wire to the headlight between the rectifier/regulator and the diode? No relay is required.

Steve

I've pulled it apart twice so I'm onto my third try now.

I'm sure I tried putting the lights in between the reg/rec and the diode but nothing happened.

I'll try that tomorrow.

It must be a half wave rectifier as that would be the reason for needing either a capacitor or a battery to make up the short fall and keep the lights bright.

[CookMySock]

It must be a half wave rectifier as that would be the reason for needing either a capacitor or a battery to make up the short fall and keep the lights bright.Well I don't really follow that logic.

Don't know, sorry bro. Beats me why that doesn't go.

Steve

[Starky307]

Well I don't really follow that logic.

Don't know, sorry bro. Beats me why that doesn't go.

Steve

Have a read through this link (http://en.wikipedia.org/wiki/Rectifier) about half wave rectifiers.

I'm off to go shopping for capacitors.

[bogan]

I would love to do this but I don't know what size capacitor to use, any ideas how to size it up?

hmm, would depend on how many pole the stator is, and whether its a full wave or half wave rectifier, and rpm, safest just to go buy a big one and see how it goes. If that doesnt work i can probly work it out on monday, away from my comp atm and on dial up :doh: so cant do now

[davereid]

The circuit you have described should work perfectly without the need for a capacitor, as a relay will cheerfully pull-in on rectified AC.

Its very unlikely that your rectifier is half wave unless it is broken !

My guess is a wiring error across the coil.

If you are using a standard automotive relay, (as stupid as it seems) the coil is not always in a "standard" place.

What is the coil on one relay, may well be the contacts on another !

Your coil will have a resistance of around 70 ohms. Look for this as your starting point.

Good Luck !

[Starky307]

The circuit you have described should work perfectly without the need for a capacitor, as a relay will cheerfully pull-in on rectified AC.

Its very unlikely that your rectifier is half wave unless it is broken !

My guess is a wiring error across the coil.

If you are using a standard automotive relay, (as stupid as it seems) the coil is not always in a "standard" place.

What is the coil on one relay, may well be the contacts on another !

Your coil will have a resistance of around 70 ohms. Look for this as your starting point.

Good Luck !

No wiring errors, I can guaranty that.

There is something going on real funny like.

I undid it all again and started at the beginning, again.

Wired the output of the reg/rec straight to the battery and put the multi meter on the battery. With the bike not running it was reading 12.56vdc, with the motor running it would go up to 14.4vdc at high rpm.

This shows the reg/rec is working fine.

Next I put the diode back in between the output of the reg/rec and the battery, once again at the battery there is an increase in voltage with the motor running but there is nothing showing on the meter if I go between the negitive connection and the reg/rec positive output (before the diode).

I really need an oscilloscope to see what is happening.

P.S I tried the capacitors but I could on get 35v 2200microF and they didn't work, if i charged them on the battery they would pull the relay in for a sec or so.

In the end I've fitted the ac regulator I got as well and have the headlight running off that, for the L.E.D taillight I built my own rectifier using 4 diodes and it all goes now, no battery needed. The only problem is the lights are dim at idle, I get I'll have to live with that for now.

Thanks for all your help guys, I've blinged you all. :) :)

[davereid]

Wired the output of the reg/rec straight to the battery and put the multi meter on the battery. With the bike not running it was reading 12.56vdc, with the motor running it would go up to 14.4vdc at high rpm.

This shows the reg/rec is working fine.

Next I put the diode back in between the output of the reg/rec and the battery, once again at the battery there is an increase in voltage with the motor running but there is nothing showing on the meter if I go between the negitive connection and the reg/rec positive output (before the diode).


You are saying that

a)..you have the neg of your meter on the battery neg
b).. on the battery side of the single diode you have 12.5 to 14.5 volts depending on RPM
c).. on the other side of the diode you get no reading on you meter, and all you have moved is the positive meter lead.

Pretty clearly this is not possible, unless the meter is being mislead ! An old analog meter would be handy here, it could be that your digital meter is unable to make a good a/d conversion on the rectified AC at the regulator output.

But it still doesn't show why the relay will pull in when connected from battery neg to battery pos, yet it wont pull in when connetec from battery neg to rectifier output.

This is only possible if the point you are placing the negative of the relay is changing, or one of your negative wires is crook.

[bogan]

this is extremely perplexing. You have tried hooking up another 12v source in place of the rectifier, this makes the circuit function as you want it too right?
Therefor there has to be some sort of problem with the stator/rectifier side of things, which there shouldn't be as it is charging the battery. All i can think of is that for some reason the stator isnt generating enough current, but this seems unlikely, do you have part numbers/tech specs for the stator and rectiier components?

[Starky307]

You are saying that

a)..you have the neg of your meter on the battery neg
b).. on the battery side of the single diode you have 12.5 to 14.5 volts depending on RPM
c).. on the other side of the diode you get no reading on you meter, and all you have moved is the positive meter lead.

Pretty clearly this is not possible, unless the meter is being mislead ! An old analog meter would be handy here, it could be that your digital meter is unable to make a good a/d conversion on the rectified AC at the regulator output.

But it still doesn't show why the relay will pull in when connected from battery neg to battery pos, yet it wont pull in when connetec from battery neg to rectifier output.

This is only possible if the point you are placing the negative of the relay is changing, or one of your negative wires is crook.

Steps A and B are right but step C I get a measured dc voltage that appears to be the difference between the battery voltage and the reg/rec output, it is around 0-1.5 ish vdc depending on RPM.

All the joints were properly soldered more than once to start with and I checked the continuity where ever possible, in the end i used terminal block connectors so I could take it all apart and check again and again what was going on at different points in the circuit.

Trust me when I say this, I have checked and re checked everything I have done over the last couple of weeks. I have taken it all apart and started again more times than I have had hot dinners this week (mainly because I was at my fathers working on my bike till late). It has me stumped, I may even try again with some proper test instruments available.


this is extremely perplexing. You have tried hooking up another 12v source in place of the rectifier, this makes the circuit function as you want it too right?
Therefor there has to be some sort of problem with the stator/rectifier side of things, which there shouldn't be as it is charging the battery. All i can think of is that for some reason the stator isnt generating enough current, but this seems unlikely, do you have part numbers/tech specs for the stator and rectiier components?

Yeah I simulated it in place on the bike by using a second battery, I used a second battery to replace the output from the reg/rec and it worked perfectly. When the second battery was connected (to simulated the bike running and the reg/rec outputting + 12vdc) the relay pulled in and the lights came on, remove the second battery and the relay dropped out and the lights stopped.

I'll get a link to the parts used tomorrow as it is all on my work computer's email.

It was all Moose racing electrical stator and regulator/rectifier.

[bogan]

Yeah I simulated it in place on the bike by using a second battery, I used a second battery to replace the output from the reg/rec and it worked perfectly. When the second battery was connected (to simulated the bike running and the reg/rec outputting + 12vdc) the relay pulled in and the lights came on, remove the second battery and the relay dropped out and the lights stopped.

right, so it has to be something to be something to do with the power source, low current, or some current looping anomoly through the relay inductor/diode (cant see how this would be possible though).

Have you tried just hooking the bulb directly to the rectifier output? would establish whether there is significant current, though i wouldnt use it like that for too long if it does light up, may wear the bulb filament out quickly

[Starky307]

right, so it has to be something to be something to do with the power source, low current, or some current looping anomoly through the relay inductor/diode (cant see how this would be possible though).

Have you tried just hooking the bulb directly to the rectifier output? would establish whether there is significant current, though i wouldnt use it like that for too long if it does light up, may wear the bulb filament out quickly

Yeah I did, when I hooked the reg/rec output directly to the light nothing happens, but if I hook the battery directly to it the voltage increases with RPM, WTF I hear you say, I've been standing next to my bike saying that alot lately.

[bogan]

yeh voltage increasing with rpm makes sense (dont know if its normal), and it cud just be the multimeter reading it a bit wrong too, actually thats probably the case.
Some pics of stator instal, rectifier, wiring, multimeter etc may shed some light on it if its easy to do. As im unsure what could be going wrong atm

[davereid]

This observed voltage increase with RPM may be (misleading) us to think the stator - rectifier combo is doing its thing correctly.

Try..
measure the charge current instead of battery voltage. That will show us if the alternator is making any juice.

Get a hefty test light- 25 watts, and put it across the output of the rectifier, testing it is actually proving the goods, not just f.n up your meter.

If no joy on those tests, put the test lamp across the INPUT to the rectifier, and see if the alternator is actually making anything.

[Starky307]

yeh voltage increasing with rpm makes sense (dont know if its normal), and it cud just be the multimeter reading it a bit wrong too, actually thats probably the case.
Some pics of stator instal, rectifier, wiring, multimeter etc may shed some light on it if its easy to do. As im unsure what could be going wrong atm

I'll try and take some photos but don't realy want to open up the bike to take a photo of the stator.


This observed voltage increase with RPM may be (misleading) us to think the stator - rectifier combo is doing its thing correctly.

Try..
measure the charge current instead of battery voltage. That will show us if the alternator is making any juice.

Get a hefty test light- 25 watts, and put it across the output of the rectifier, testing it is actually proving the goods, not just f.n up your meter.

If no joy on those tests, put the test lamp across the INPUT to the rectifier, and see if the alternator is actually making anything.

I didn't have access to a meter that I could measure the charge current of the battery but I'll see what I can do this week.

Check out this link (http://procycle.us/dskits/lightcoils.htm)as it is the items i have.

I brought both the ac headlight regulator and the 12v Regulator/Rectifier, note it states it must be conected to a battery to function properly, this is the big catch with the whole system.

As mentioned previously I know the stator works as the headlight is now working using the ac regulator and the LED tail light is running of my own bridge rectifier. See pic attached.

[Starky307]

Pic attached this time.

[bogan]

It does seem odd that the 12v regulator rectifier needs a battery, and a capacitor wont just do the job. But it sounds like using the ac regulator makes it work as intended so youll just be running with that now i assume?

[bogan]

just read from another link that the voltage regulator requires a reference voltage to work properly, ie a 12v battery, an when the diode is in place that voltage reference cannot get back to the rectifier, so this could be your problem.
Though having a regulator that requires an external reference voltage makes very little sense to me, i may even try to find a reg schematic to figure out why this might be, and whether it is the most common approach.

[Starky307]

It does seem odd that the 12v regulator rectifier needs a battery, and a capacitor wont just do the job. But it sounds like using the ac regulator makes it work as intended so youll just be running with that now i assume?

Yeah I'll keep the ac regulator in it for now, it just makes the lights dim at idle, I wanted to overcome that by having the battery, I guess I'll have to live with it as is, god lifes tough.

I thought of just using the bridge rec I made to do the job of the reg/rec i brought but then I wanted to know the theory for the reg/rec and my wiring not working.

If i get bored over the next week or so I might try it as I have drawn to see if it would work. See pic

[bogan]

Yeah I'll keep the ac regulator in it for now, it just makes the lights dim at idle, I wanted to overcome that by having the battery, I guess I'll have to live with it as is, god lifes tough.

I thought of just using the bridge rec I made to do the job of the reg/rec i brought but then I wanted to know the theory for the reg/rec and my wiring not working.

If i get bored over the next week or so I might try it as I have drawn to see if it would work. See pic

yeh thatd prolly work (ac reg i assume ur using) but the charging of the battery may be a bit iffy, as it wont be getting a regulated voltage/current.

sounds like a bit dim at idle shouldnt be too much of a problem, ur not often needing high visibility at idle!