VTEC and ignition timing?
Joined: Nov 2009
Posts: 677
Likes: 5
From: south florida
I'd like to hear how Brian does it too, but I'll tell you how I do it. After doing a very good fuel tune for a very flat AFR, I do several "good" street dyno pulls, and average them together, and compare the shape of the torque plot to the injector pulse width. In areas where the torque curve goes down while the injector pulse width goes up, I add some ignition timing (maybe 1 degree). Then I do more street dyno pulls. If the dip gets smaller, I add more timing in the same area, otherwise, I go the other way with it. Once my fuel and torque curves match as good as possible, I start moving the whole timing line up or down to find the maximum curve. If I start getting knock counts somewhere, I start altering the slope of the ignition curve, and continue. Once I hit peak torque or hit the knock limit, I back off a bit.
Former Sponsor
Joined: Mar 2002
Posts: 15,526
Likes: 19
Thanks for sharing your method of street tuning, that's great information. How close do you think this gets to a dyno tune? Also how do you tune partial throttle?
How much knock count do you consider too much, and how do you alter the slope of the curve? Once you hit peak torque or knock limit, do you back off the entire curve, or just that spot and keep going with the rest?
How much knock count do you consider too much, and how do you alter the slope of the curve? Once you hit peak torque or knock limit, do you back off the entire curve, or just that spot and keep going with the rest?
If the car generates knock counts that don't go away by retarding timing, then I would accept those as ghost knock. However, if there is no ghost knock, then I don't want to see any knock counts at all once the tune is done.
Former Sponsor
Joined: Mar 2002
Posts: 15,526
Likes: 19
Originally Posted by gernby' timestamp='1333466458' post='21571054
The dynamic conmpression is lower on the high cam than it is on the low cam, so optimal ignition advance is automatically increased because of that. Ignition timing on the high cam is less dependent on RPM compared to the low cam, but very dependent on volumetric efficiency.
Also, what affect on fueling does lowering VTEC have, I'd guess it leans out the mixture, which should also add power? Is it likely that lowering VTEC alone without tuning for it would cause a loss in power?
BTW, I don't believe there's any benefit to going lean. I think that's just a misunderstanding caused by VAFC type piggy-backs that lean AFR while advancing timing. People assumed that they were making power because they were making them leaner, but they were really making power because they were advancing timing.
Registered User
Joined: Jul 2007
Posts: 1,376
Likes: 2
Gernby, do you have any suggestions as to how AP1 users can adequately monitor knock? I picked up an E-Manage Ultimate not too long ago and have been cautiously making small changes. The fueling is straight forward, as expected, but I'm having a difficult time finding a reliable way to detect knock. Thus, I am hesitant to make timing changes if I can't adequately detect detonation. The EMU can log the knock sensor signal, but the sample rate is only 50Hz, which is too slow in my opinion to capture a knock event.
Your comments about comparing injector pulse width and torque values is interesting, but not applicable to AP1s due to the limitations of the com port. Any other tricks we (AP1 owners) can try to use for determining optimal timing? One approach I'm considering is to spend some time on the dyno with 108 octane in the tank in order to find MBT for my engine. When running on pump gas, I would revert to a more cautious timing tune.
I have traditional knock head phones, as well as a knock mic I built. Neither solution has allowed me to hear detonation, unless I haven't found the timing limit yet. In an attempt to induce a bit of knock to see if I could hear it, I was adding timing while the engine was on the low cam. Distinguishing between knock and what was just engine noise was nearly impossible, in this scenario. I was adding up to 12 degrees of timing, which I would think would be enough to cause knock at WOT between 3800 and 4500 RPM on the low cam. My valvetrain lash is within the factory tolerances, so I don't think it's a case of abnormally high engine noise.
Thoughts, suggestions? Thanks.
Your comments about comparing injector pulse width and torque values is interesting, but not applicable to AP1s due to the limitations of the com port. Any other tricks we (AP1 owners) can try to use for determining optimal timing? One approach I'm considering is to spend some time on the dyno with 108 octane in the tank in order to find MBT for my engine. When running on pump gas, I would revert to a more cautious timing tune.
I have traditional knock head phones, as well as a knock mic I built. Neither solution has allowed me to hear detonation, unless I haven't found the timing limit yet. In an attempt to induce a bit of knock to see if I could hear it, I was adding timing while the engine was on the low cam. Distinguishing between knock and what was just engine noise was nearly impossible, in this scenario. I was adding up to 12 degrees of timing, which I would think would be enough to cause knock at WOT between 3800 and 4500 RPM on the low cam. My valvetrain lash is within the factory tolerances, so I don't think it's a case of abnormally high engine noise.
Thoughts, suggestions? Thanks.
Former Sponsor
Joined: Mar 2002
Posts: 15,526
Likes: 19
Gernby, do you have any suggestions as to how AP1 users can adequately monitor knock? I picked up an E-Manage Ultimate not too long ago and have been cautiously making small changes. The fueling is straight forward, as expected, but I'm having a difficult time finding a reliable way to detect knock. Thus, I am hesitant to make timing changes if I can't adequately detect detonation. The EMU can log the knock sensor signal, but the sample rate is only 50Hz, which is too slow in my opinion to capture a knock event.
Your comments about comparing injector pulse width and torque values is interesting, but not applicable to AP1s due to the limitations of the com port. Any other tricks we (AP1 owners) can try to use for determining optimal timing? One approach I'm considering is to spend some time on the dyno with 108 octane in the tank in order to find MBT for my engine. When running on pump gas, I would revert to a more cautious timing tune.
I have traditional knock head phones, as well as a knock mic I built. Neither solution has allowed me to hear detonation, unless I haven't found the timing limit yet. In an attempt to induce a bit of knock to see if I could hear it, I was adding timing while the engine was on the low cam. Distinguishing between knock and what was just engine noise was nearly impossible, in this scenario. I was adding up to 12 degrees of timing, which I would think would be enough to cause knock at WOT between 3800 and 4500 RPM on the low cam. My valvetrain lash is within the factory tolerances, so I don't think it's a case of abnormally high engine noise.
Thoughts, suggestions? Thanks.
Your comments about comparing injector pulse width and torque values is interesting, but not applicable to AP1s due to the limitations of the com port. Any other tricks we (AP1 owners) can try to use for determining optimal timing? One approach I'm considering is to spend some time on the dyno with 108 octane in the tank in order to find MBT for my engine. When running on pump gas, I would revert to a more cautious timing tune.
I have traditional knock head phones, as well as a knock mic I built. Neither solution has allowed me to hear detonation, unless I haven't found the timing limit yet. In an attempt to induce a bit of knock to see if I could hear it, I was adding timing while the engine was on the low cam. Distinguishing between knock and what was just engine noise was nearly impossible, in this scenario. I was adding up to 12 degrees of timing, which I would think would be enough to cause knock at WOT between 3800 and 4500 RPM on the low cam. My valvetrain lash is within the factory tolerances, so I don't think it's a case of abnormally high engine noise.
Thoughts, suggestions? Thanks.
Man, I wouldn't even try to give advice about your situation. I believe I've mastered '06+ AP2 FlashPro tuning (NA only), but my whole process depends on great high-res data from the ECU.
Registered User
Joined: Mar 2011
Posts: 354
Likes: 0
nmrado, couldn't we use a similar method to Gernby's tuning based on torque graph off a software dyno? Were you able to datalog the pulls while adding timing, did the changes make more power? Also, why would you tune for MBT using 108 octane rather than pump gas, unless you plan to run that while racing? (not STR legal?)
Registered User
Joined: Mar 2011
Posts: 354
Likes: 0
BTW, I don't believe there's any benefit to going lean. I think that's just a misunderstanding caused by VAFC type piggy-backs that lean AFR while advancing timing. People assumed that they were making power because they were making them leaner, but they were really making power because they were advancing timing.
Former Sponsor
Joined: Mar 2002
Posts: 15,526
Likes: 19
I haven't noticed much change in power from changing AFR, so I just tune AFR for a balance between efficiency and safety. Since the safe AFR is really RPM dependent, I just tune so that it's close to stoich at low RPMs, then slowly gets richer as the RPMs rise, finally hitting about 12.5 at 8K.
Here is a thread I did about using the DBW throttle map to help with part throttle tuning. It's not going to work for non-DBW cars.
https://www.s2ki.com/s2000/topic/789...on-the-street/
Here is a thread I did about using the DBW throttle map to help with part throttle tuning. It's not going to work for non-DBW cars.
https://www.s2ki.com/s2000/topic/789...on-the-street/
Registered User
Joined: Jul 2007
Posts: 1,376
Likes: 2
What do you mean by a restricted throttle? Also how much resolution/sample rate would you say is needed for logging knock?
nmrado, couldn't we use a similar method to Gernby's tuning based on torque graph off a software dyno? Were you able to datalog the pulls while adding timing, did the changes make more power? Also, why would you tune for MBT using 108 octane rather than pump gas, unless you plan to run that while racing? (not STR legal?)
nmrado, couldn't we use a similar method to Gernby's tuning based on torque graph off a software dyno? Were you able to datalog the pulls while adding timing, did the changes make more power? Also, why would you tune for MBT using 108 octane rather than pump gas, unless you plan to run that while racing? (not STR legal?)
Legend, do you have an AP1? If yes, you should be aware that the OBD2 port uses an older protocol (ISO 9141-2, if I remember correctly) which communicates much slower than the '06+ AP2 ECUs. It's not very usable for logging...
I'm thinking about using 108 octane gasoline in order to reach MBT without detonation. This would tell me what the upper bound of my timing curve should be. Timing values above that would just be needlessly increasing cylinder pressure while not producing any extra power. It's possible to find 100 octane pump gas every now and then, but I don't want to run a 100 octane tune all of the time. So, I still need a reliable method of detecting knock to establish a regular pump gas timing tune.
Joined: Nov 2009
Posts: 677
Likes: 5
From: south florida
I'm thinking about using 108 octane gasoline in order to reach MBT without detonation. This would tell me what the upper bound of my timing curve should be. Timing values above that would just be needlessly increasing cylinder pressure while not producing any extra power. It's possible to find 100 octane pump gas every now and then, but I don't want to run a 100 octane tune all of the time. So, I still need a reliable method of detecting knock to establish a regular pump gas timing tune.