12V AC motorcycle regulator

Discussion in 'The Projects Forum' started by Suzukiman, Jun 11, 2010.

Thread Status:
Not open for further replies.
  1. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    Hi,
    I am new here and have been lurking on and off for quite a while.

    I need a regulator that regulates AC to 12 and/or 6V . Some older off road motorcycles have a lightning coil that runs the 35W headlight off AC and without a regulator the bulbs just blow regularly.

    Later models were fitted with regulators that is just ground on one side and the other lead to the AC feed from the Alernator/magneto. Thus a shunt type regulator.

    Would it be possible to use two SCR's in parallel and how can one contol them?

    If anyone knows of an existing schematic I will really appreciate a link.
    I have attached a schematic without the triac control for any advice.

    Thanks
     
    Last edited: Jun 13, 2010
    charanjit singh likes this.
  2. oidium45

    Active Member

    Apr 24, 2010
    130
    8
  3. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    The DC side with battery and charging coil has its own regulator and that is already sorted out.

    In addition I have the single wire from a seperate coil in the magneto the runs the headlamp bulb with AC and blows them regularly. Later models had a very basic 12 V AC regulator which is grounded to the frame and one wire to the AC wire.

    The method seems to be briefly touched on in the link you provided. It could be a triac between the AC wire and ground then switched on with a zener when the AC voltage rises above 12-13V or should a diac be used to trigger it? It has a very low component count and is very small.

    I will post a diagram a bit later on how I suspect it is regulated if that would help.

    Thanks for the reply.
     
    charanjit singh likes this.
  4. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    Hi,
    I have attached a schematic for the 12V AC regulaor that I need to the first post. I will appreciate any advice on how to control the TRIAC to shunt any AC voltage above 12v AC to ground. Will a Zener work?
    Thanks
     
  5. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I know you were looking for an SCR/TRIAC application, but I don't happen to have any SCR/TRIAC models loaded at the moment.

    Anyway, have a look at the attached. A TRIAC application would be somewhat similar. I used an N-ch MOSFET instead.

    The output might not look very smooth, but it averages 13.9v, which is typical for an automotive system.
     
    Suzukiman likes this.
  6. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    SgtWookie,
    Thanks that would definitely work, but I should have mentioned that I have one of the OEM regulators and they do not put out DC, but AC in and out, as its only for the headlight and has its own dedicated winding in the magneto. There is then an additional DC charging circuit with rectifier/regulator at 12v and battery for the horn, flashers and rear lamps with its own windings in the magneto.

    The OEM is unfortunately potted, but actually pulls the AC rail down to 12 -14 volts AC. It is very small about 1" x 1" and has a very low component count.

    I appreciate your help so far.
     
    charanjit singh likes this.
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well then, maybe someone else has a circuit for you.

    [eta]
    Keep in mind that if you use a TRIAC (or a pair of SCRs) that once the TRIAC/SCRs are turned on, the only way to turn them off is to stop the current flow through them.

    I suspect that rather than using SCRs/TRIAC as shunts (which would clamp the gen output to 0v when they are turned on) that they are used in series to charge a cap on the output; once the cap reaches a certain voltage the SCRs/TRIAC starts getting switched on later and later during the input sine wave.
     
    Last edited: Jun 13, 2010
  8. Solcar

    Member

    Jun 8, 2007
    21
    3
    SgtWookie has a good idea about how the regulators are often done with the control of the turn-on delay.

    I wonder how the shunt circuit you have is so small. I would think it would get pretty hot, too. I played around with one on the LTspice simulator without really being able to replicate the magneto coil. Using a 60 hz transformer instead, i was able to get something that pretty much worked but had bad high voltage spikes.
     
  9. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    As I understand it, the actual AC regulator is very crude and its only function is to ensure that the AC voltage does not go over 12 - 14v if the RPM's go up and prevent the headlamp bulb from blowing. I am not sure if spikes are managed at all and would not think so. It only drives the headlamp bulb which is very forgiving and works with AC or DC, in this case AC. The current is limited by the design, amount of turns and diameter of the windings in the alternator and can be shorted out and not damage the windings. I have just removed a dead short on the AC feed to ground that has been present for a long time on a bike and the AC supply is still working, just too high +- 26V and blowing the bulb.

    Most people confuse this with the normal DC circuit which is still present on these bikes. In actual fact there are usually 3 power supplies coming from the alternator/magneto on these off road bikes. One for the headlights using AC, another for the charging circuit to the battery, rectified to DC and regulated, and a third which powers the CDI and allows the bike to start from a kick starter and run even if lights and battery are removed.

    In actual fact the regulator I have just fitted to a Honda XL250 has a heatsink and is bolted to the frame. It gets quite warm with the lights on, but not very hot. The frequency of the AC would be quite high as the engine can go up to a good couple of thousand RPM. I am not sure if it shunts out a half wave of full wave.

    On the bike forums there is quite a history around the 80's on that headlight system and for a while honda used a 3.81 ohm 30W resistor to shunt the AC output on the dim light, later they providd a regulator and better bulb and moved from 6V to 12V all round.

    I would appreciate any suggestions or advice.
     
  10. Solcar

    Member

    Jun 8, 2007
    21
    3
    I did more LTspice simulations and came up with a triac circuit which looks like it might work. The Zener diodes are 15v 500mw types. Since the shunting kicks in based on peak voltage, you might have to choose Zener values up to 50% higher, or more, than 12v. The resistor affects the trigger point some, too, but going too high could make the circuit operation erratic.
     
    Suzukiman likes this.
  11. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    Thanks a million, that looks like what I need. I will order some parts and see how it works. It may take a while from my parts supplier as it takes up to two weeks to get an order delivered here. I will try and post results as soon as I can.
    Thanks again, your feedback is really appreciated.
     
    Solcar likes this.
  12. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    Solcar, the parts have arrived and I will pick up tomorrow and will post as soon as I get some time to build and test.
     
  13. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    I received the parts to day and put together, but it pulls down the AC rail to 1.8V. Here is the schematic I used. Any help will be appreciated.
     
    Last edited: Jun 22, 2010
  14. Jack_K

    Active Member

    May 13, 2009
    115
    0
    I guess I must be totally missing the boat but it sounds like he wants to rectify the AC to 12 or 13 VDC. Why not just a diode and cap into an LM317 set to 13 volts or something similar?

    Jack
     
  15. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Jack,
    An LM317 would burn up. It's a 35W headlamp. That's around 3A current - and the bulb is burning out.
     
  16. Jack_K

    Active Member

    May 13, 2009
    115
    0
    OK. Just buffer it with a transistor. I didn't mention that since I thought it would be obvious.

    Sorry for my short sightedness.

    Jack
     
  17. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    Jack,
    Would you mind giving more detail as I am a bit electronically challenged.
    Thanks
     
  18. Jack_K

    Active Member

    May 13, 2009
    115
    0
    I don't have a schematic drawing program here. Go to National's web site and look at the data sheet and app notes on the LM317 and other regulators. Google for National Semiconductor.

    Jack
     
  19. Suzukiman

    Thread Starter Member

    May 1, 2010
    94
    4
    Jack,
    Thanks for the input, but I am looking at a simple solution which has a low component count and thus will not require a complex schematic and large complex enclosure, very similar to the OEM AC regulators.
    I have attached a picture of the Honda OEM and the Aftermarket AC regulator to show the small footprint.
    The green on the OEM and the Black wire on the aftermarket is the ground wire. The Yellow and/or White just "T"'s into the AC wire from the alternator to the headlight. It is not in series with the wire. This confirms that it must be a shunt regulater and not series.
    The so called OEM "crude" regulating is what I am trying to replicate if possible as it has been proven to be fairly robust, but these units are just very hard to obtain in our part of the world.
    I really appreciate your input.
     
  20. rjenkins

    AAC Fanatic!

    Nov 6, 2005
    1,015
    69
    From what I remember of my motorcycle repair days, the 'regulator' was nothing more than a pair of big (stud mount) zener diodes connected in series, back-to-back.

    With suitable heatsinking, a pair of 1N3312B diodes (50W each) should work fine.

    It would limit the AC at around 13.7V which is well within the normal '12V Battery' regulated range, so bulb life should not be a problem.

    If it's connected directly across the bulb, so only dissipating power while the lights are on, the actual power dissipation should not be too high as it's only taking the excess above what the bulb uses.
     
Loading...
Thread Status:
Not open for further replies.