Ignition Fix part 2

Discussion in 'The Projects Forum' started by SilvrEclipse, Dec 26, 2007.

  1. SilvrEclipse

    Thread Starter Active Member

    Nov 21, 2007
    58
    0
    Ok guys, The first circuit Ron made works flawless, but now I need help constructing the other side of the circuit.

    Background - The first circuit we made converted a ground pulse signal to a 12vv pulse. The pulse ran through the emanage computer and then to a transistor pack which grounded the coils. The problem now is that the transistor pack wont work with this application. The car will idle perfect but when the turbo kicks in the car skips and misses. The transistor pack gets real hot so I believe this is the problem.

    What I need - We will be inputing a 12v 12mA pulse signal. This will need to complete a ground from the coil pack to the battery. Not sure how much current travels through this ground but I assume it is a couple of amps. The resistence from the 12v in to the ground is .6 ohms. Could this be accomplished with a single transistor?
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Sounds like what's happening is that the spark isn't "hot" enough to jump the plug gap under the higher density of fuel/air mixture.

    What's your plug's spark gap, and what coil(s) are you using?

    Up to the early 1970's, practically all vehicles used a .035" gap for plugs; but this was the days before electronic ignition. Secondary ignition voltage was around 22-25kV, and was not sufficient to bridge a greater gap when the engine was running with open throttle. The narrow spark gap was a problem at idle; the stratified charge was tough to light with that small of a gap.

    Electronic ignition changed that. The newer systems had a much higher secondary voltage; in the range of 40kV. The spark gap was widened to provide much more reliable ignition at idle; for some vehicles the gap went out to .080". There were some "teething problems" with secondary wiring and distributor cap/rotor materials; anyone who had a mid-70's GM V8 powered vehicle will attest to having the rotor short out at 40,000 mile intervals due to burn-thru. :rolleyes: Changing from bakelite to a different plastic formulation corrected that.

    OK, but I digress - bear with me, I do that frequently ;)

    If you are feeding just ONE pulse to the coils, your spark will not be very intense. Additionally, if you are using transistors, you will be dissipating a LOT of power across the transistors themselves. A typical PN junction has a voltage drop of 0.6V to 0.7V across it; as it saturates, much more is dropped. You would be served better by using power MOSFETS, like an IRF1407. This little jewel will handle 75Vdss, has an RdsON of 0.0078 Ohms (!) and can handle a drain current up to 130 Amperes :eek: (properly heatsinked, of course)

    Rather than feeding the gate a single pulse, I suggest using a gated oscillator circuit of around 25kHz. This will rapidly switch the MOSFET on and off, causing the coil(s) to issue multiple sparks. Use a non-polarized high voltage capacitor across the coil(s) to absorb the back-EMF. Try using around .0033uF for starters.

    [eta]
    Forgot to mention - you should be using components with an industrial temp rating or better. Commercial rated parts won't survive long under your vehicle's hood.

    [eta]
    Lucas came up with a rather dodgy solution for the Jaguar 5.3L V12 ignition system. Keep in mind, this was quite a few years ago and transistorized ignition systems were in their infancy. The Lucas system had two ignition coils which were connected in parallel; one of which had the secondary tower plugged (!) with what looked like Lucite, a clear plastic. This 2nd coil was located in front of the radiator, far from the "real" pimary coil. The transistor(s) were located in the valley near the primary coil at the rear of the engine. A reluctor with 12 points was on the distributor shaft, it's passing was detected by a small pickup coil, which triggered the transistor(s) to provide a pulse to the coils. Since one of the coils had its' tower plugged, there was no place for the secondary voltage to go; when the field collapsed it caused a high reverse voltage to be applied to the other coil. It was basically a resonant tank circuit; the plugs would fire multiple times due to this 2nd coil with the plugged tower.

    The Lucas ignition was replaced by the Marelli ignition system in 1989 - which wasn't much of an improvement. :rolleyes: Quite a few 1989 and later XJS V12 Jaguars were lost due to fires caused when one half of the Marelli system malfunctioned, causing raw air/fuel to be dumped into a hot catalytic converter.
     
  3. SilvrEclipse

    Thread Starter Active Member

    Nov 21, 2007
    58
    0
    Thats exactly what I think is happening. The gap is set to .028, when I run the ignition the stock form it works perfect but when I try to run it through my built circuit I get problems.

    I believe you are trying to make this system to complex, the system only needs to fire one time to ignite the fuel. Anything more is just over kill. My system runs off the stock computer so if the ECU fires the plugs multiple times per revolution then so will this. I just needed to find something that could drain the coils. I may try the IRF1407 and see if that helps.
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I just read through your other thread; I hadn't seen it before.

    What's the original maximum boost for your engine in stock configuration?

    Did you modify the engine to lower the compression ratio, like adding a 2nd head gasket?

    Seems to me that American manufacturers limited the max boost to around 6PSI using a waste gate. But you're talking about 16PSI boost! Normal atmospheric pressure is about 14.7 at sea level; the US manufacturers' standard boost bumps that to around 20.7PSI. You're going to 30.7PSI. If your engine is still stock, you're going to have extremely high cylinder head pressures. Don't be surprised when you start breaking things. It's no wonder you're having ignition problems.

    To get the most out of adding more boost, you must reduce the compression ratio of the engine. Of course, this will negatively impact the power and fuel economy you'll get when running on normal atmospheric pressure, but you can't have everything. By retarding your spark, you're limiting the amount of power that's produced - but until you drop the compression ratio of the engine, you don't have much choice; it's retard the spark or detonation will occur.

    The least expensive way to reduce the compression ratio is to use two head gaskets. Not the best way, but the least expensive.

    You could also use different low-compression pistons, shorter connecting rods, or have the crankshaft re-ground (in decending order of preference).

    You might need to go to surface-gap type sparkplugs.

    You may need to use hotter coils.

    If your secondary ignition wires are less than perfect, you'll have problems. Using copper conductor secondary wires may help, but you must then use resistor-type spark plugs - or else your ignition system will be a big Tesla spark generator, and you will be generating broadband RF noise.
     
  5. SilvrEclipse

    Thread Starter Active Member

    Nov 21, 2007
    58
    0
    This motor was NOT boosted in stock form. It has been turboed for over a year on 10psi with not problems. With complete stock ignition the car runs perfect so I know everything works fine. But when I add my build circuit it hits 2psi and starts to stutter.

    The motor is getting replaced with a built one in a few months.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    OK, so try the cheap things first ;)

    Go for an IRF1407. Actually, install a couple in parallel, two for each coil, on a couple of heatsinks. Copper heatsinks are about twice as effective at removing heat than a comparably sized aluminum heatsink. Use heatsink compound to maximize heat transfer, but the most important thing is a good solid physical connection. Keep your leads from the IRF1407 to the coils and ground short in length to minimize added inductance/capacitance. Using a coaxial cable for your gates wouldn't be a terrible idea; stray inductance from ignition cables or other sources might be enough to trigger your gates at the wrong times.

    These MOSFETs shouldn't even get warm. However, under extended operating conditions in high temps, they might. MOSFETs have an unusual characteristic; as they get warmer, they increase in resistance, unlike other semiconductors. If you have a couple of them in parallel and one starts getting warm, it will conduct less, thus generating less heat, and have a chance to cool off a bit. This unusual property makes them a snap to run in parallel. Transistors aren't so easy.

    Although these IRF1407's are theoretically rated for 130 amps continuous, the TO220 package has a physical limitation of 78 amps. That's still far more than you need, but why use a 2x4 when a 10x12 will work just fine? ;)
     
  7. SilvrEclipse

    Thread Starter Active Member

    Nov 21, 2007
    58
    0
    ok man thanks for the info, I will let yall know how it works. Might be a while though, radioshack doesn't seem to have them so I guess I will have to order them.
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
  9. SilvrEclipse

    Thread Starter Active Member

    Nov 21, 2007
    58
    0
    Ok I pick up a IRF510 transistor from the store just to see if it would work. I couldn't even get the plugs to fire with this setup. The 510 transistor works the same as the 1407 but the 1407 is just more heavy duty.. Right? Any reason why the 510 would not at least work for this?
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    It's not a transistor; it's a MOSFET. The IRF510 is in the same family as the IRF1407, but if you compare the specifications, it's like the 510 is a 1967 VW beetle and the IRF1407 is a Lamborghini Countach.

    Check to see if the output from your controller module is reading 0V to ground when the coil is not being triggered.

    If not, you might need a simple differentiator circuit to drive the MOSFET's gate negative to turn it off.

    Recap - you're using the SOURCE pin to switch the ground to the coil, and the DRAIN pin connected to ground, and the GATE pin connected to your controller module's output, right?
     
  11. SilvrEclipse

    Thread Starter Active Member

    Nov 21, 2007
    58
    0
    Yes that was how I had it connected. I need to trace down the problem more. The only problem I can think of is that the MOSFET cant drain enough power from the coils to fire the plugs. I guess I have some more experimenting to do...
     
Loading...