Spark Plug Gap Experiments and Extremes
 

I've been hunting a misfire at idle and am suspecting my injectors or coils/plugs. This led me to some interesting videos about plug gap which I figured others might find interesting.

TLDR, both of these videos show that increasing spark plug gap (all the way up to 0.100) increases power output. The amount seems negligible (3-5 HP on a NA LS), but the results seem to be consistent. Anecdotally, I've found people mentioning that running a larger gap had a positive effect on drivability as well. More power and better manners? As long as the spark isn't blowing out I don't see a downside.

My car is down right now while waiting for some parts, but I'm going to increase my spark gap up to 0.045 and see if the misfire goes away. I figured someone else might find these interesting. I'd also be curious to hear what coils and gap people are running at various boost levels.

VVT coils, 330hp, .020-.022 gap. I'd try decreasing idle AFR target to the ~13.5 range, that'll tell you if it's injectors. Nothing wrong with 13.5, just means they're a bit shitty in the smaller pulsewidth range.

Curly, I'm probably going to make a thread about my idle woes after I've done a bit more to try and correct it. In the middle of winter I couldn't idle much above 12.5 without getting a lot of stumbling.

I opened the gap on my plugs and (maybe unsurprisingly) my car began misfiring at high boost (~200kpa). I gapped them down a bit and tried again but was still getting some misfires.

Recently, after reading a handful of threads about them, I turned the dwell on my R8 coils down. This made me wonder how spark plugs gap relates to dwell. I assume if you crank dwell up close to the max for the coil, then you have more energy and can run a larger spark gap.

I was running ~1.6 Ms of dwell, but have bumped that up to 2 ms now. I'm leaving the plug gap at 0.035 and going to see if the misfires go away.
​

Coil issues will almost never rear their head at idle, however injectors will almost 100% of the time show their weaknesses at idle. I would send the injectors out, or look at your injector settings.

Sure enough, bumping the dwell up a bit seems to have fixed the WOT misfires. I might try bumping the gap up to see if they'll take 0.040 or 0.045 with the added dwell.

Curly, I assume you mean "send them out" to have them cleaned and tested?

Forced induction is different than NA in regard to sparkplug gap. Factory turbo vehicles even run smaller gaps. I do not recommend anything over .032".

Leo, is that with stock coil packs? I've always heard to gap down spark plugs when going to forced induction, but I'm still under the impression that a stock or larger spark gap (to an extent) shouldn't be an issue with upgraded coils.

Gap down is correct for FI because of the increased cylinder pressure. Even with high output coils/ignitions FI still requires a smaller gap relative to NA.

We still run .020-.022 with upgraded coils on Miata’s and BMWs. B58 supra coils are supposedly the bees knees for turbo BMWs, still .022”.

Thanks for the info guys.

Curly, on your Miata's are you running nicer (platinum, iridium, etc...) plugs?

Yes. I'd look at small pulsewidth data and other idle settings first though.

Nope, just BKR-7Es gapped to .020-.025ish.

Can you post your tune here? Might be a drop box thing, since Link files might not be supported.

Curly, tune is attached. I'm using FMS 1050cc injectors. I emailed them and am using their recommended settings for small pulse width.

I have been experimenting with allowing CLL to run at idle, and have tried idling at a couple different target ignition angles (10, 12, 16). The car seems to be idling better now that it's warming up, but it still misses here and there and likes to idle very rich.

Your target is 12.94, what's it running at when it's "very rich"? Got a log?

Biggest thing I see is my CLL gain is about x10 yours, mine is 10, 15, 30, 30, 30, 30, yours is 1, 1, 2, 4, 8, 10. You might try my settings and see what happens. otherwise you have plenty of min/max trim, so there isn't any reason for it to run rich. Now if you're CLL is at -15% when warm and you're still running rich, maybe there's a heat soak issue or warm up enrichment issue. You're pulling 5% fuel when cold, maybe you need to pull more and decrease base fuel table values. Again, depends on what the CLL is doing when running "very rich".

I meant to attach a tune yesterday but couldn't figure out how to export part of a tune unless it's in CSV format (IDK why PCLink doesn't let you chop up a log file).

I've attached 2 here, one from today with CLL enabled at lambda and one from about a month ago with CLL disabled (and much colder ambient temps).

I'll look into the gain table. I changed the update rate at idle today from 1 to 3. CLL seems to be helping, but I think it's also the warmer ambient temps. The car idles a tad bit rougher at 13 compared to being richer, but it's not stumbling. It's strange too because sometimes it'll idle around 14-15 smoothly, but that's usually after a cold start or hot restart. I'm still figuring out all the closed loop stuff, but it's pretty handy. On my old Speedy I did open loop everything because I couldn't get the closed loop to work properly. It's refreshing to turn on a closed loop algorithm and have decent results with pretty minimal setup/troubleshooting.

I spent more time figuring out that cold IAT enrichment than I'd like to say. These injectors seemed to be pretty bipolar around freezing or below. TBF I don't think I ever ran the FlowForce 640's in the middle of winter, so I'm not sure how those would have fared.