L98 Spark timing and E38 knock sensitivity

[CV8IIM6]

Hi All,

I recently had my 09 VE SSV L98 A6 on the dyno for a power run and noticed what I suspect may be false knock retard. Vehicle is only ever run on 98 octane and is completely unmodified (aside from a relocated VY IAT sensor). The knock only seems to appear on the dyno and the octane scaler was at 0.00 prior to the dyno runs.

http://i449.photobucket.com/albums/qq213/CV8IIM6/logknock.jpg

http://i449.photobucket.com/albums/qq213/CV8IIM6/SparkTable.jpg

I run what I believe to be a conservative high octane spark map yet still logged 3 degrees of KR during the run. I know timing should be dialed in using MBT on a dyno but I’m looking for some educated opinions before I go down that route.

I thought I should be able to get away with a couple of degrees more timing at WOT? And, I’m thinking about desensitizing the knock sensors by about 10%. Opinions appreciated.

Cheers.

[GMPX]

I would suggest that some tuners at this point would listen to the engine (literally), you are right though, the final timing figure you had doesn't seem over the top.

[CV8IIM6]

I logged a few more WOT runs over the weekend (not on the dyno but still in a responsible manner :angel_innocent:) and noticed a similar amount of knock retard.
logging individual cylinders i found some were more prone to knock than others at times but they all seemed to be too sensitive.
I've adjusted sensitivity at key areas of each map by 10% but have not had time yet to log any more data.
Hope i'm on the right track.

http://i449.photobucket.com/albums/qq213/CV8IIM6/KnockThreshold.jpg



I would suggest that some tuners at this point would listen to the engine (literally), you are right though, the final timing figure you had doesn't seem over the top.

I haven't had any experience setting up knock boxes or knock phones but might have to check out a few options. Anyone know of any units that are affordable for the home tuning enthusiast?

[gmh308]

I logged a few more WOT runs over the weekend (not on the dyno but still in a responsible manner :angel_innocent:) and noticed a similar amount of knock retard.
logging individual cylinders i found some were more prone to knock than others at times but they all seemed to be too sensitive.
I've adjusted sensitivity at key areas of each map by 10% but have not had time yet to log any more data.
Hope i'm on the right track.

I haven't had any experience setting up knock boxes or knock phones but might have to check out a few options. Anyone know of any units that are affordable for the home tuning enthusiast?

How you travelling with this?

Also, how accurate did you find the ECM engine torque number against the dyno torque measurement?

[Highlander]

Ross is there a way to output the Knock actual raw data on the scantool?

Thanks

[CV8IIM6]

How you travelling with this?

Also, how accurate did you find the ECM engine torque number against the dyno torque measurement?

Hi gmh308,

I've done a lot of 1st and 2nd gear WOT runs and adjusted all the knock tables where necessary by 10% 5% then 2% at a time until knock was no longer evident. I then did a couple of 3rd gear runs and noticed a little bit of KR, I suspect due to the additional load on the engine. At this point I decided the dyno would be a much safer and more efficient place for this kind of activity so things have been on hold ever since.

Hopefully I'll have some dyno time within the next month or so. I think I'll disable the knock sensors and adjust timing to achieve max power/torque. Then re-enable the sensors and trim the knock sensitivity tables as necessary.

I logged torque in Nm and the dyno was set up for ftlb. When I take an average (eg between 70kmh and 170kmh) the engine torque and the dyno torque are very close (within 10ftlb). At peak torque however, the dyno reads nearly 30ftlb lower than calculated engine torque.

Cheers.

[CV8IIM6]

I’ve been doing a lot of data logging lately setting up a MAFless tune and have continued trimming the knock sensitivity tables running total timing as above. I had to increase the value of some of the cells by up to 40% in order to eliminate most or the false KR.

I also managed to get a little more dyno time the other day, I played around with WOT timing and mixtures with interesting results. I had about 3 runs on a mainline dyno (L98 Auto in 4th gear), this is what I found:

I had a bit of a freak run with the stock standard tune. It averaged 12.0 AFR, knocked the whole way through the run resulting in average spark of 18 Deg and still made 219rwkw.

The next run with my tune maintained 12.8 AFR and 20 Deg making 220rwkw. So it seems with similar timing, leaning the mixtures from 12.0 to 12.8 made NO difference.

I then added 2 Degrees which leaned the mixtures to 13 AFR. Audible detonation was evident, timing was pulled and power dropped to about 213wrkw.

At this point I ran out of time.

Next time I’ll try 22-23 Degrees total timing again and aim for 12.5 AFR to see if I can improve on 220rwkw. But, at this stage it appears that my L98 is at its limit on 98 octane fuel with standard intake and exhaust and won’t take much more than 21 Degrees total.

Cheers.

[RetroAus]

We achieved around the same with bone stock VE on Mainline .....

[L31Sleeper]

What is a 2009 L98 (I'm only familiar with the L98 350 in the C4 corvette)

[gmh308]

What is a 2009 L98 (I'm only familiar with the L98 350 in the C4 corvette)

An LS2 6L bottom end with L92 heads and car intake (now superceded to LS3 intake - same/similar head casting as LS3, L76, L99).

Fundamentally the same plant as in G8 Pontiacs.

Out of GM they are L76 - with AFM, L98 - without AFM.

They emerged for MY06. :)

[The Alchemist]

Hers one of my tables with a cammed SSVE we did recently....some knock tables need no changing wheras some need quite a bit of work to eliminate false knock....

5731
varies from engine to engine.......

5732

Heres a standard cammed SS VE map for you to compare what you're doing to. See how the timing at peak torque ,4400 rpm, pulls out then increases again as rpm climbs to the limiter. The Blue area is unused or not accessed with standard engine.
On our 95 octane aim for 22 to 23 deg between 3200rpm to 3600rpm, 18 deg by 4400 at full load and about 24 degs by 6200rpm. :secret:
See how you go.
Cheers,
Mike

[gmh308]

Hers one of my tables with a cammed SSVE we did recently....some knock tables need no changing wheras some need quite a bit of work to eliminate false knock....

Heres a standard cammed SS VE map for you to compare what you're doing to. See how the timing at peak torque ,4400 rpm, pulls out then increases again as rpm climbs to the limiter. The Blue area is unused or not accessed with standard engine.

On our 95 octane aim for 22 to 23 deg between 3200rpm to 3600rpm, 18 deg by 4400 at full load and about 24 degs by 6200rpm. :secret:
See how you go.
Cheers,
Mike

How are you differentiating false knock from real knock?

[The Alchemist]

Best torque with lowest timing on a dyno and knock phones :)

[hymey]

I found most e38s have different sensitivity from car to car, When I tune I have them relaxed and if i here detonation on the dyno it also shows on the KR map. It is how i find knock but also determining at what load point it occurs. And MBT occurs a fair bit b4 knock sets in, I generally see 5hp loss after MBT b4 detonation shows up and at that time the knock retard is pulling timing but not soon enough. I then pull timing back until power drops off then bump up two degrees, then put a slightly more aggressive KR map in that will show its ugly head on the dyno if knock is a problem, once u find MBT u can leave the knock settings sensitive as u are 2-4 degrees from knock anyway.

For eg a cammed VE I did today I put up to 32 deg timing in it and it still refused to knock on 98 pump fuel.Pulled retard of 0.1 deg with relaxed settings then and could here onset of detonation but not bad. Quite typical, But hp dropped by 5 at 32 degrees, go back to 30 increases by 2, back to 28 increases by a further 3. Keep going back and then move up I ended up with 22 peak torque 28 full power, Which gave consistent results at 12.8:1 afr. With even sensitive settings in the car it still didnt knock on street or dyno and I new I was 4 deg away from knock and with sensitive settings it would never be an issue.

I big thing I found was that on certain settings I have in KR I can actually find MBT, and still be well clear of real knock, one i find it i relax them a little. a good track tuning trick.

[CV8IIM6]

On our 95 octane aim for 22 to 23 deg between 3200rpm to 3600rpm, 18 deg by 4400 at full load and about 24 degs by 6200rpm. :secret:

Thanks for the top secret tip :cheers:

[gmh308]

Best torque with lowest timing on a dyno and knock phones :)

Dirka dirka! Ta!

[odd boy]

2 days ago we did 2007 lumina SS "similar to VEss" cammed 588/588 228/228 114, long sureflo headers, and air intake only, I do 25 deg with 95 RON octane, no knocks showed up. We got around 420 hp and 400 tq on my dyno (dynomite). Little bit conservative.

Some facts:

I do admit “these cars love richer fuel and advance timing as well!!!!!”
These engines make 4 deg retard while they are stock with stock tune!!!!!
Their knock sensors give fault readings and they are too sensitive.

[CV8IIM6]

Thought it was about time i reported back on this topic.

I pulled a lot if timing from most of the high spark map and still had to do a lot of work on the knock threshold tables to eliminate the false knock retard recorded in the logs.

On the road the car feels much stronger everywhere.

On Dyno Dynamics stock made 220rwkw.
With shielded pod filter, x-force twin 2.5" exhaust, x-force stepped headers with cats and of course the modified tune made 260rwkw.

I've attached my current modified tune and the original tune for comparison, comments welcome.

cheers