Whippled 496

iTrader: (4)

not sure how to flop the X and Y axis but here is how to convert units....

dc_justin

iTrader: (13)

In my experience, burst knock is much needed with FI... I've tried reducing it on mine, and quickly see HUGE knock spikes as well as audible knock with rapid throttle transitions... EG, 35-100% throttle stab. Then again, I've only done that test with CA 91 Octane (equivalent to about 83 octane anywhere else... )


As far as Jimmie's tune goes, the knock recovery rate won't really help... I've seen a bunch of his logs that show that the knock is persistent, aka sawtoothed. You can't really tell much though because it is so frequent... aka, 5.0, 4.9, 4.8, 5.0, 4.9, 5.0, knock after knock after knock, which says two things immediately.

First is that the amount of timing being pulled is not sufficient for the amount of knock being detected.

Second is that with it consistently picking up knock like that even with almost no timing and very good gas, there is something else going on... serious engine noise, or (hopefully) improperly filtered knock sensor...

dc_justin

iTrader: (13)

That, you can't do... the Log configuration is what I was talking about (since he posted up log shots).

All units are configurable (metric/imperial, AFR, IFR, etc), but in the tune tool it's not configurable.

If it really bugs you that much, you can highlight the whole table, copy with labels, switch to the default window and choose "Conversions" tab. Hit the Create new table button, then hit transpose, and it'll flip the axis for you and you can view it there...

UNLESS (and this is a big UNLESS) GM had no other "official" way for the TCM and ECM to communicate and the only way it can cause spark to retard is via altering the knock sensor output. Much in the same way whipple's electronics play with the IAT and other sensor outputs... Not likely at all, just presenting it as a very unlikely option.

With everything we've considered already... my money's on the filter, or something banging...

kbracing96

iTrader: (26)

What about changing to a different OS. Maybe there is a bug in the o3 year os that is causing it to see false knock. Just an Idea???

dc_justin

iTrader: (13)

He mentioned at some point he has a 2 bar sensor waiting to go on, but wants to get this resolved first, so I'd be willing to bet that is in the plans.

On the topic of OS's, with you running COS 3, do you have any problems with your OL ECT Multiplier not having it's intended effect? I see no effect from that table causing me to be super lean for a few minutes at startup. EFI Live folks looking into it...

kbracing96

iTrader: (26)

No, mine seems to be working fine, just looked a log I did this morning and it appeared to work just as it should. I am running MAF open loop right now though.

02sierraz71_5.3

iTrader: (2)

I personnaly like having the knock recovery scaled down it helps alot with being able to see the actuall cell that causes the knock with the stock settings for kr it will pull it out forever and it causes bleed over into cells that arent causing knock and dont need to be changed, I reduced my stock value by 70% meaning my timing recovers 70% faster and it only bleeds over into the adjacent 1 cell instead of 5 cells.

-Adjust your pe enrichment rate to max your running lean when you first hit wot.

-I dont know enough on the filter setting I would consult with someone from EFI live it is very possible they know something we dont.

-I would be shooting for 12:1 afr if your too rich it will cause knock like dewanshu mentioned

-Others mentioned adjusting the VE table, I would bump the WOT part of the table to a value of 130 across the board (looks like a table top) and leave it reason for this is because that is the only time you are using it since the maf is still enabled (meaning at wot your seeing a mix of ve and maf). Before doing this log your maf hz at wot and let us know if the maf is maxing out, at the top of first what is your maf hz reading?

-Set your pe lambda value to .816 across the board, then tweak the maf curve so that your afr (wideband) equals 12:1 the maf curve has alot more resolution than the pe fueling table this is why I like doing this, plus when you get your 2bar map and OS going then your pe table values are where they should be and then you use your ve table to line the fuel up.

Now we know that his kr is false its time to answer why, Im still thinking on it.

dc_justin

iTrader: (13)

Assuming the filter is the culprit, it is possible that it simply wasn't filtering out normal engine noise... could be valvetrain, pulleys, whipple rotor noise transferring through the mounting bracket, etc, etc.

02sierraz71_5.3

iTrader: (2)

Im hoping its that simple, but the logs he just sent us have kr and they are labeled as 4.7 filter? He could desensitize the sensors alot but Id do that as a last resort.

another strange thing is in the logs cruising around at a steady state and all of the sudden it will pull 20 degrees of timing map, tps, no kr it shouldnt be pulling timing and it is?
heres a pic

dc_justin

iTrader: (13)

That timing dip is probably the AC compressor cycling. I typically see around 20* of timing pulled as the compressor cycles.

One important thing to note is that with that tune he used with the 4.7 filter, he has NO IAT spark correction. That is absolutely a contributor to his knock. The CAM2 log has, on average, higher IATs than the first log. So even with the higher octane, he's still got too much timing for the IATs.


And for the most part, his fueling is dead on, within 2% of commanded AFR for 95% of the log. Logged PID CALC.BEN_LC11 shows this. Map it out via RPM/MAP and check out the averages. 0.99 - 1.02 across the majority. No real need to adjust VE table.