Not the usual 555 question

Discussion in 'The Projects Forum' started by Man_in_UK, May 13, 2008.

  1. Man_in_UK

    Thread Starter Senior Member

    May 13, 2008
    132
    0
    Can anyone please help ?

    I think this project can be sorted with a simple 555 circuit but I can't handle the calculations and there might be a better way ... so all good ideas greatly recieved.

    I want to build a circuit that will reduce the power of a bike engine when a trigger is sent. My thought was to place a timer in between the coil pickup sensor and the electronic ignition. When I need to power to be reduced a very small delay from the timer will retard the timing and cause the power to drop.

    Easy I thought. I can get as far as guessing that the normal rev range is about 4000 RPM and the pickup will be giving out 66 pulses per second ... how to turn this into an ignition retard of about 5 to 10 degrees is WELL beyond me.

    All suggestions / diagrams / tissues to cry into are welcome
     
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    There is probably a way to do the ignition delay, but it would be helpful to understand the point of the project.

    Let me expand on that - if this is to cut power for shifting, then there are better ways. If this is for RPM limiting, there are probably better ways.
     
    Last edited: May 13, 2008
  3. Man_in_UK

    Thread Starter Senior Member

    May 13, 2008
    132
    0
    I have seen a safety device used at wheelie schools to cut the engine if the bike angle gets too great. I have heard that these can cause problems as the power comes back at full power once the angle has gone back down. I was thinking of building a similar thing only I want to reduce power and not lose it completly.

    Its going to be on a single cylinder bike so I do not have the option of cutting some of the spark plugs.
     
  4. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    66 pulses/sec comes out to 15 milliseconds between pulses. A 10 deg delay is on the order of 420 microseconds. That is doable.

    The fun part is now to understand the nature of the pickup pulse. What generates it, and also what does the ignition module need to trigger it? That is necessary information before proceeding.

    Doing this "right" would entail a variable delay that adjusted with RPM. In this case, a fixed delay might be best. Your scooter engine can probably do better than 4 grand - my Honda Fit is good for 6,500 at red line. But the fixed delay will be felt more strongly (perhaps weakly is better) at higher RPM's. Plus it's much simpler and cheaper to build.

    Come up with the ignition pulse information. By that time, somebody else may have had a better idea.
     
  5. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    It is probably easier to do what you ask with a small microcontroller. Here is a link to a homemade CDI for model airplanes. Your ignition may not be CDI, but the program for the controller is open source and includes a retard/advance routine. At least you can see one way to do it effectively. There are at least two other DIY ignitions (all CDI) on the RCUniverse Engine Conversions forum as well as commercial units (see: CH Ignitions) available. John

    http://www.modelbouw.gompy.net/newcdi/newcdi.htm
     
  6. Man_in_UK

    Thread Starter Senior Member

    May 13, 2008
    132
    0
    Im not sure of exactly how in depth info your talking about but the trigger is sent from a fixed coil that picks up a magnet on the flywheel (but you know that already).

    Or have I got to get a scope out and give you EVERYTHING ?



    I did think about using a PIC microcontroller on this. I can program them up and get them running fine on a bench but as I usuall cheat/miss out a good reset circuit they never seem to stay reliable for me. It is worth a look at the code mind.
     
  7. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    It would be important to know the polarity of the trigger pulse, the necessary pulse height, and if the ignition triggers on the rising or falling edge of the pulse. If a device has to go in between the pulse generator and the ignition module, it is necessary to know how to make an acceptable trigger pulse from a new source.
     
  8. Man_in_UK

    Thread Starter Senior Member

    May 13, 2008
    132
    0
    I had not even given thought to the trigger being the falling edge. Now I will have to work out a way to scope the trigger against the spark ..... I don't even know how I'm supposed to do that.

    They do say a little bit of knowledge is dangerous, I'll hold me hand up that that one.

    How do I check if it triggers on the falling edge ?
     
  9. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    If it helps, I think most of the model aircraft electronic ignitions trigger on the falling edge. I would suggest turning the crank fairly slowly by hand and looking at the trigger. If you have a two-channel scope you could look at the trigger from the sensor and the trigger for the spark on separate channels. For model CDI, the spark trigger is just an SCR and it is quite easy to see where it is being triggering in relation to the pick-up, which are mostly Hall sensors today. Older engines used coil sensors.

    John
     
  10. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    The other check is to set up a 555 on a slow cycle, about .5 Hz or less. If the pulse out of pin 3 triggers the ignition module, the we will know that a 12 volt pulse will do the job (you will attach it to the bike battery for power). As the pulse will change at such a long interval, a meter on pin 3 plus an observation of when the spark is produced will demonstrate triggering on the leading/falling edge.
     
  11. Man_in_UK

    Thread Starter Senior Member

    May 13, 2008
    132
    0
    A new way thinking gave me another idea. Is it a good or bad one ?

    The first idea relies on me duplicating the pulse from the coil pick-up and then having the ability to delay it(to recap i want to reduce the power & not remove it completly).

    What about if I leave the pick-up to send its pulses and just interfere with them ?

    If we have 66 pulses per sec(half are unused as its a 4 stroke = 33 that do something). Can I strap a transistor across the coil and pulse it 10 - 20 times a sec.
    Can I assume it will remove some of the pulses and leave some unchanged ?
    (this should make the engine run badly but still run)
     
  12. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    I am not sure I understand. You have a 4-stroke at 3960 rpm (66 rpsecond). Power strokes are 33 per second for a single cylinder.
    Now, you want to control the power with a pulse rate of 10 to 20 pulse per second. That will end up with a very syncopated firing and rough engine, if it runs at all. There will be periods when you are almost in-sync with the exhaust stroke, hence no power; other periods when you are on compression near TDC, with power; and periods when you are so out of sync that you will have knocking and back-firing.

    I question whether it will work. Although, that method of control sounds a little like the hit-and-miss engines of the 1800's (http://en.wikipedia.org/wiki/Hit_and_miss_engine).

    My advice is to use the ignition to make the engine run well; use the throttle to control power. John
     
  13. Man_in_UK

    Thread Starter Senior Member

    May 13, 2008
    132
    0
    No, that was not what I mean.
    Not replace the correct timing pulse but disturb it enough to lose power.

    I do not know how well it would run if I was to remove every other firing stroke, I can only imagine that with the engine at about 3k revs it might feel like a misfire.

    An ideal circuit would be a controller that could count every timing pulse (on compression) and remove every second or third one to create an engine that runs with a bad misfire.

    Rather than do the job correctly I was going to just strobe a pulse at a fixed frez and see what happens. What I'm not sure about is the correct way to short out the origonal pulses and what effects that might cause.
     
Loading...