Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Motorcycle 12VAC to DC Headlight Wiring circuit

Status
Not open for further replies.

Techedz

New Member
Hi, 1st post so be gentle!

I'm trying to convert a 12VAC circuit to 12VDC, but the trick is, it has to output High and Low on an LED Headlight using the original switch. I'm trying to build a 'plug and play' unit to replace the current weak AC Globe. The LED draws from 1.5 to 4 amps depending on which one I use.

I've drawn a rough circuit of what I'm thinking but not sure about the eatrh/AC side? Note there are two bridge rectifiers, one is to convert the higher current for the light, the other just converts a couple of hundred milliamps to switch the relay.

Thoughts? Am I on the right track or miles off?

XR ac - dc Headlight Wiring circuit.png
 
You are on the right lines, but you need two changes.

1) the negative of the little bridge that feeds the relay must only connect to pin 86 of the relay. It must not connect to the negative of the main rectifier.

2) You need two more 6A diodes, or another rectifier. One diode has to have its anode connected to AC (High) and its cathode connected to the +ve of the capacitor. The other diode has to have its cathode connected to AC (High) and its anode connected to the -ve of the capacitor. If you use a rectifier, connect its AC connections to AC (High) and AC (Earth) and its +ve and -ve connections to the capacitor.

If you make those changes, I think it will work.

Item 1 is needed when low beam is on to prevent the relay being energised through the AC (Low) connection when that is negative compared to AC (Earth).

Item 2 is needed when high beam is on to give a path for the +ve and -ve high currents to the capacitor from AC (High)
 
Good points Diver300, I think I also made an error, I don't think there is power from the low when switched to high. I'll go and check on my mates bike and redo the circuit with your suggestions. Cheers Ed
 
I've drawn out the circuit that I described.

Headlight_relay.png


You can get three-phase rectifiers like this:-https://uk.farnell.com/multicomp/sbr2506/diode-bridge-rect-3-ph-25a-600v/dp/2750988

A three-phase rectifier has three ac inputs, an you would simply connect those to AC(High), AC(Low) and AC(Earth). The + and - outputs of the rectifier would go to the capacitor. That would give the circuit exactly as I have drawn it.

If you use two single-phase rectifiers, you can connect AC(High) and AC(Earth) to the ac inputs of one, and AC(Low) and AC(Earth) to the ac inputs of the other. The + and - outputs of both are connected together and to the capacitor.

That would not be quite the same as I have drawn. Using two single-phase rectifiers like that would have two diodes in parallel in two places. The ones labelled D2 and D3 would be duplicated, but that wouldn't matter and would reduce the voltage drop slightly.
 
Thanks Diver300, that's all great info.

So I can use one three-phase rectifier and a small cheap single-phase rectifier to switch the relay?

BTW, did some testing today with a 6A single-phase rectifier initially. Had to use the cap to stop the light flickering and the 12.7V input voltage jumped from 10.7 to 11.3V on the DC output.

But then I swapped in the proper Chinese motorcycle single-phase rectifiers with heat sink that I bought for $14 and the voltage jumped to 11.5 and the headlight was brighter. It must have a built-in capacitor because there was no flicker without the external cap and when I connected the cap, the voltage dropped, the light dimmed and the amps went up!

The link you had sells the three-phase rectifiers for $26 ea without a heatsink, I found some 50A 3phase 1000v on eBay for $11.28 local with a heatsink, $7.56 from China (3-6 weeks post generally) so I might buy a local one to test it and if it works, order more from China. They are the same price with/without a heatsink.

Also checked temperatures today, the small 6A single-phase rectifier jumped to 50C with a small aluminium plate bolted to it, the motorcycle one only went to 40C. That was worse-case scenario with a 60W (5A) bulb to test it.
The LED lights draw a lot less power, they range from a best of Hi/Low of 7/18W to max 25/43W (2/3.23A) so the rectifiers only got to 25-35C. Definitely need the heat sink, the data sheet shows at 100A + the amperage drops towards zero at 150C. Looks like I won't get anything near that. I'll draw that up and post it to check I understand what you're saying.

Not quite sure on the wiring where you mentioned using the two single-phase rectifiers? If you connect them both together on the output, I assume you would still need the small cheap single-phase rectifier to switch the relay between high and low? That would be good because I already have half a dozen single-phase rectifiers here and on order so I can use them up.

The easiset way to do all this really, is to chop out the existing 4 wire AC regulator on the bike and replace it with a single-phase rectifier because it goes from there to the headlight switch. However, I'm building these LED light/ac-dc kits for people with very little mechanical or electrical knowledge so I'm trying to make a plug-and-play unit that plugs onto the headlight wiring after the Hi/Low switch to simplify installation.
 
I drew this for the 3 single wave rectifiers from your diagram and instructions. Does it look right?
I'll draw up the 3 phase one as wellView attachment 139093.
That is correct.

However, if you connect the second ac input of the second 6A bridge to AC(Earth), it will halve the current in the D2 and D3 of the first 6A bridge, so it may keep the rectifiers cooler.

Wiring circuit diver300.jpg


Using two single-phase rectifiers is often easier than using separate diodes because they are easier to mount. Also single-phase rectifiers are very cheap. I have used bridge rectifiers where only two diodes are needed for both those reasons on separate occasions in the past.
 
Last edited:
It's very difficult to know why the different rectifier, lamp and capacitor combination took more or less current. If you have an oscilloscope, some traces might help. The LED lamps are designed to run on steady DC, so their behaviour on rectified ac, with or without a capacitor is difficult to predict. The LED lamps have elaborate circuitry inside, and it seems that your motorcycle regulators are complicated as well. It all makes analysis far more difficult than for rectifiers etc where the behaviour is well known.

I can't see any advantage of running incandescent lights using the rectifiers.
 
It's very difficult to know why the different rectifier, lamp and capacitor combination took more or less current. If you have an oscilloscope, some traces might help. The LED lamps are designed to run on steady DC, so their behaviour on rectified ac, with or without a capacitor is difficult to predict. The LED lamps have elaborate circuitry inside, and it seems that your motorcycle regulators are complicated as well. It all makes analysis far more difficult than for rectifiers etc where the behaviour is well known.

I can't see any advantage of running incandescent lights using the rectifiers.
The incandescent lights run fine on AC, the problem is the light is poor and they change with engine revs because there is no battery. They are running off the alternator only getting AC.

I imagine the LEDs all have different circuitry inside them but they are rated to handle 9-24v. So far, they seem to handle the rectified AC. I have a nice little digital meter on the output to check voltage drop and it shows 100Hz which I thought was strange as it means they are still getting some sort of AC signal? However, the light output is very clean and smooth. I do some more testing but I'm busy with other things until next week. I don't have an oscilloscope and don't really have a strong enough electronics background to read one but I have a friend who is retired who has one and knows how to use it so I can take it all to him to test later next week and document it all with the different regulators. Cheers Ed
 
Just out of curiosity, will the relay work on AC or does the frequency get so low that it drops out?

Mike.
P.S. Well worked out Diver on the reverse flow in post #5.
 
Well I'm quite confused by all this :D

Mostly by points labelled 'earth' - obviously with the suggested connections shown, the 'earth' on the headlamp must be removed - if indeed there's an 'earth' connection there.

But I'm an ex-biker, and have had a number of trail bikes with this kind of electrical system, usually 6V (back in the day) - assuming it's what I think it is?.

The alternator is just two coils underneath a spinning magnetic 'dish', one end of both coils is connected to chassis (earth), the low power coil connects via a half wave rectifier, and charges the (far too small) battery, this runs the indicators, brake light, side lights etc. The other (high power) coil feeds just the headlight, and is a specially designed coil that limits it's voltage/power to prevent the bulb blowing - so at a certain number of revs the headlamp reaches full brightness, and doesn't get any brighter, as revs increase further - but at low revs the headlight is next to useless (as often are the indicators, brake light etc.).

I thought long and hard about improving it, but could never come up with a suitable solution, mostly because of the common chassis connection of the coils - and as it was my sole means of transport, I didn't really have the time to mess about with it.

Obviously now, with the availability of LED headlamps, it's a LOT easier, as current requirements are MUCH lower - but I'd be wanting to add a battery, with a decent charging circuit, to avoid the changing voltage issues completely.

And what's with the relay?, and where's the dip switch? - adding a relay doesn't seem to give any advantage?, and won't have been used on the original wiring.
 
And what's with the relay?, and where's the dip switch? - adding a relay doesn't seem to give any advantage?, and won't have been used on the original wiring.

My interpretation is that the block labelled "Headlamp switch" is the original three wire connection to a dual filament headlamp bulb, so it has two switched power wires for full beam and dip beam, plus return.

The circuit is an inline adapter to allow replacement of the (AC) filament lamp with a three terminal DC LED type lamp.

The relay is needed to switch power to the appropriate LED lamp terminal, after the AC from whichever feed has been rectified.
 
My interpretation is that the block labelled "Headlamp switch" is the original three wire connection to a dual filament headlamp bulb, so it has two switched power wires for full beam and dip beam, plus return.

The circuit is an inline adapter to allow replacement of the (AC) filament lamp with a three terminal DC LED type lamp.

The relay is needed to switch power to the appropriate LED lamp terminal, after the AC from whichever feed has been rectified.

Right - so why not just put a bridge on the input to the Dip/Main switch - so just a single bridge and capacitor needed, and no relay.
 
My interpretation is that the block labelled "Headlamp switch" is the original three wire connection to a dual filament headlamp bulb, so it has two switched power wires for full beam and dip beam, plus return.

Using the correct bulb will eliminate much of the problem. A LED bulb for both high and low beams in one bulb would allow the original switch and wiring to be used. Just a diode bridge to convert the alternator output to DC is all that is added to the circuit that way. Just a single bridge before the switch.
 
The "Earth" connection on the LED headlight is just the negative connection, and is unlikely to connect to the frame of the bike.

The advantage of the circuit suggested is that the whole lot can be mounted into the headlight housing. It may be difficult to get to the wire that goes from the alternator to the dip switch. There may also be a headlight on-off switch in the way as well.

Most alternators have an inherent current limit, due to the inductance of the windings. With a fixed magnetic field, if the speed doubles, the emf generated by the alternator doubles, but so does the impedance of the inductance. At most engine speeds the emf is significantly larger than the output voltage, and the impedance of the inductance dominates the total impedance.

I had a motorbike with a 6 V system, and a 25 W headlight and a 5 W tail light, so a total of 30 W or 5 A. I changed the headlight bulb to a 12 V, 55/60W headlight, plus a 5 W panel light in parallel with the dip beam. That didn't need any changes to the alternator coil or the main/dip circuit.

Car alternators are a bit different, in that the field winding allows that magnetic field to be changed. At high current, as with the bike alternators, the impedance of the inductance dominates. If the load suddenly reduces, there is suddenly no voltage drop caused by the impedance and the voltage will be very large until the magnetic field reduces.

That is the cause of load-dump surges in car alternators.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top