Can't keep the motor from detonating!

[tech3sandcars]

I am having some trouble trying to figure out why I can' t stop this motor from denonating, MS2/Extra purchased from DIYAutotune about a month ago, here is the background:

1) Motor is 3.0 Subaru, Low impedance injectors, 800cc (ya a little big but I have run them before). PWM is setup properly I think. I have tried different settings but makes no difference.
2) 36-1 wheel, LS Truck coils, configured in waste spark
3) I have mechanically checked TDC using a long reach plug and rotating the motor around till it stops up against the plug in both directions and found the exact center between the 2.
4) My timing gun matches the what the MS is thinking
5) motor fires up and runs great, idles well, and runs to about 4000-4500rpm or so without a hitch. Pulls strong, runs smooth.
6) motor will denonate once we start past that RPM. I have a handheld knock meter (can pick up quicker than my ear) but I can also here it once in a while
7) I have backed timing clear down to 30deg BTDC I have run these motors at 36 and 38 before with no detonation.
Running Tunerstudio with autotune turned on, purposely started fat and can see tunerstudio lowering the fuel numbers.
9) when it denonates AFR is as fat as 12:1 and timing is at 30deg.
10) AFR gauge is an AEM Wideband unit. The gauge matches what we are seeing on the computer screen.

I have posted the MSQ and a zip file of the datalog, can someone please have a look and comment on the MSQ to start. I think there may be an issue with the high cc injector and such small duty cycles and injector on times and the resolution of having a small ReqFuel may be getting in the way.

Thanks in advance, Gary

[Matt Cramer]

You can attach the log as a ZIP file. This will tell us more about what is going on - air temperature, for example.

[tech3sandcars]

Got it, Datalog now included in first post.

Just to add a little bit more data, the motor will free rev past my point of the problem I have at 4000-4500rpm, the detonation occurs when I am trying to use the VE table tuning in tunerstudio (which is set to normal). Maybe I should go back to manual tuning.

[elutionsdesign]

Don't really see anything in the logs in that rpm range, you do have an instance of fuel cut at 6000. Have you checked the plugs, specifically examined the insulator for metal bits? Detonation can breed detonation once you get a bunch of small bb's of melted aluminum all over the combustion chamber. If its detonating at what you think is very safe ignition timing and AFR then I would start looking for mechanical reasons.

[tech3sandcars]

Thanks for looking, the Rev limiter is set for 6k so that's probable why you see that, I did look at plugs, they looked OK, they are new so not much history on them but I did not get out my magnifying glass for closer look.

[billr]

I gotta ask... is your timing light a simple one, or dial-back; have you checked timing with the light up to full rpm?

What fuel, what compression-ratio?

[xr4x4ti]

What is your dwell set at? Somebody on the YellowBullet mentioned that the LS truck will "autofire" if the dwell exceeds something like 5ms which would completely through off your timing (advancing it).

[vw_chuck]

Are you sure it is actually detonating? Do you hear it detonating or are you basing it on what the MS3 is putting out?

[tech3sandcars]

Yes the timing light is a simple one, good old school craftsman, my current dwell I believe is 3.5ms, the motor is bone stock, might have 9.0:1 compression, its the EZ30D Subaru 3.0 from the early 2000's. I am using a handheld knock meter that uses the Subaru sensor so the tuning for the sound and frequency is for that motor. The meter picks it up before I can but I have heard it also a few times.

I have run ms2 on 2 other motors like this with no issues, the only difference is the low Z injectors and the LS truck coils on this one.

Anybody have comments about the low ReqFuel value compared to the numbers in the VE table compared to the size of the injectors? It seems to idle just fine and normally that is more of an issue with large injectors.

[kaeman]

in general settings, you have incorporate afr in the settings, but you have the ego disabled. could that be giving you some problems.

[tech3sandcars]

Good question, I am assuming it gets ignore if it's disabled but that is just a guess, anyone have a comment on this?

[elutionsdesign]

I'd believe if it were a problem you'd see it in the logs. These LS truck coils, are they the ones with the heat sink? If so I've heard of many tuners testing and refusing to use them due to spurious uncontrolled firing.

[jsmcortina]

Have you checked the timing near the RPMs where you are getting a problem?

James

[tech3sandcars]

Yes these are the coils with the heat sink, I have checked the timing with a timing light free revving it and it seems to be OK, will check it again. Seems odd that these coils would misfire, they don't do it in the oem vehicle, is it a dwell issue?

[jsmcortina]

Seems odd that these coils would misfire, they don't do it in the oem vehicle, is it a dwell issue?

Your dwell settings looked ok, but the "random advance" issue is certainly worth investigating.

James

[kjones6039]

These LS truck coils.......................spurious uncontrolled firing.

I use the LS coils and there are a lot more out there that don't seem to experience this problem.......

My $.02 for this morning!!

Ken

[tech3sandcars]

Might be an easy swap out right now, I may try that. I have been doing lots of reading on this and my dwell should be very safe but I may decrease to 2.5ms like some guys have done.

[dontz125]

I've heard of many tuners testing and refusing to use them due to spurious uncontrolled firing.

From the testing and research some of the folks here have done, it is neither spurious nor uncontrolled. It is a deliberate current cut by the controller once a certain dwell limit is exceeded - IIRC, the limit is ~4.6 ms. Past that, the current is chopped, which of course fires the coil. If you had 5ms of dwell set and it fired at 4.6ms, assuming 6000rpm that's 9° of unexpected - and quite likely undesired! - spark advance.

I may be misremembering the limit; it may change under different conditions. It is a definite 'Gotcha!' that you need to be aware of.

[tech3sandcars]

Thanks for that comment, I formed a theory last night while on an airplane home from a business trip. This configuration is wasted spark so the coils are firing twice per cycle, so I think it even compounds the problem. My current dwell is set at 3.1 but when firing twice as often the coil thinks the rpm is twice what it actually is. I'm gonna drop the dwell to about 2.2-2.5 and see if the problem disappears, otherwise it's off to the junk yard to try the LS1 or LS2 coils.

[billr]

Wasted-COP should be no problem at the rpm you are talking about; except, perhaps for coil heating. Even 12,000 rpm allows 5 msec for each (wasted-spark) ignition event, plenty of time for the dwell and spark duration..