No announcement yet.

LS2 / LS7 Coil Packs - Issues & Observations

  • Filter
  • Time
  • Show
Clear All
new posts

  • LS2 / LS7 Coil Packs - Issues & Observations

    I posted this on KA-T, but I figured more input = better:

    I just upgraded my car to LS2 coils, which apparently require more dwell time than the LS1 coils (as per AEM wizard). I've read they're both ground triggered (0-5v to initiate dwell, remove current for spark), and that they're just the same as LS1 coils with more spark intensity (per MegaSquirt Sequencer Coil page). I've also seen a writeup of someone using LS2 coils on an RB, and under the assumption that Nissan ECUs operate similarly, LS2 coils should work on my car.

    However, I'm getting some pretty massive spark blowout. Unfortunately, my wastegate won't keep boost under 10psi, so I was stuck gapping down plugs to get a decent run while autocrossing last weekend. These things were breaking up slightly at 10psi with the plugs gapped to around 0.024", which used to run fine on the stock coil at above 0.030". So, now I'm looking back at what to do about my coils.

    Here's my theory:
    The stock KA distributor (and I'd guess most Nissan COP (coil on plug) setups) uses a single igniter. To keep from frying it at 100% dwell, you run it at 66% - 75% duty cycle max (I read one person state 44% dc is the magic number). On the AEM EMS, this is your Dwell Max setting. Since the stock dizzy has to fire 4 times per revolution (each revolution of the CAS disk equaling two engine revolutions), and there are 24 teeth, you're firing every 6 teeth. Thus, the stock setting of 4.5 Max Dwell on the AEM keeps the coil from constantly staying on.

    Applying this to the LSx setup, you're now firing 4 individual coils. Each coil should, in theory, have 24 teeth for the CAS to go through before firing again. Thus, when converting to individual coilpacks, shouldn't Dwell Max also be changed to accommodate the longer available charge time?

    Some rough calculations:
    At 4000rpm, with 24 teeth per rotation, and 1 min = 60000ms, you're going through:
    4000 rev/min * 24 teeth / rev = 96000 teeth / min * 1 min/ 60000ms
    = 1.6 teeth / ms or 0.63 ms/ teeth

    Dwell max @ 4.5 teeth at 4000rpm = 2.81ms
    Dwell max @ 18 teeth at 4000rpm = 11.25ms (completely unnecessary, but allows the full 5.5ms charge)

    And then near redline (because who stops at 6500 anymore?):
    7000 rpm @ 4.5 teeth = 1.61ms
    7000 rpm @ 18 teeth = 6.43ms (still allows a full 5.5ms charge)

    Someone please let me know if I'm way off-base here and should just buy LS1 coils like everyone else, or if I'm actually getting at something useful. Also, sorry for the long winded post. I'm just running this all through my head trying to figure out possible solutions before next week's test & tune.

    As an update:

    I verified that the dwell setting on the LS2/LS7 GM# 12573190 coil packs are being controlled by the AEM EMS. Basically i turned dwell down really low and obtained the expected results: complete spark blowout at less than 5psi. With dwell up near 6ms things went back to "normal", and I was able to hit 10psi with small hesitations from spark blowout. Also, running the AEM's Dwell Max to 15 teeth had no noticeable effect (good or bad) so far. I also verified that startup calibrations on 4-coil cars allow for 100% coil DC (such as the EVO IX at 4 teeth Dwell Max for 4 Ignition teeth).

    The next thing I'll check is plug wires, to see if the crimped down NGK's aren't making a strong enough connection (ordered some new connectors and wires from Summit), as well as looking over my ground wires. So far the engine feels great other than small-ish hesitations at high boost near peak torque (primarily in the 4k to 5k rpm range IIRC).
    Last edited by AceInHole; 06-01-2009, 05:41 AM.

  • #2
    Did you ever solve your problem here? I don't remember if you did or not. I know you posted the same thing on KA-T. just curious to know if i decide to step in that direction


    • #3
      It turns out I was just getting some timing drift. It seems as though timing was retarding something like a rev limiter at higher RPMS, and combined with pulling timing under boost, it was pulling enough to feel like a rev limiter. So far I've tweaked the settings enough to get by boosting at 10psi, but I'll definitely be working it out over the next week or so.