Trouble keeping pulse width high enough at high RPM

[koabi]

http://vems.hu/vemstune/sharingcenter/reports.php?cmd=view&key=QVzj2r

log is above, fun starts at 2:40.

General background info:  350 small block chevy with large cams and 14:1 compression.

You will notice VE values are extremely high at and above 6000 RPM.  Despite that, the injector pw decreases which causes the car to go lean and misfire.  I haven't been able to figure out why, which is why I am posting here.  The only strange setting I've found thus far is injector effective rampup which is at 1008us.  If I lower this to the recommended 200us what effects should I expect?

Thanks for looking!  Let me know if other details are necessary

[z0tya]

Interesting VE lookup (interpolation problem???) at this time (one tick before VE lookup 174%, and now 148%):

[GintsK]

4ms coil charge time is OK here? You have issue with low dwell at high rpms! Try 1 in coil charge MAP scale.

My recommendation is to switch to 1.2.0 because of much nicer Lambda sampling. I guess issue has something to do with oxygen at lambda sensor. Poor burning, misfires or something like that. With 1.2.0 you will see way better.

VE extremely high?  1200cc? You have req fuel about twice lower as calculated (5700/8/1200*6.5=3.85). So we should expect VE around 160...180. But you have just 100.... Do you have some dampening device in fuel lines? 1200 do hydro hammering a lot and it can disturb VE table.

Your cams has big overlap. I would not recommend to mess with speed density. At such overlaps engine loosing VE rapidly when throttled. Nothing wrong with it in case if you have clean MAP signal. But you have not...

Also take a look on your acc enrichment dTPS curve. I think first value in it (89) should be 0 or something close to 0. In some cases you have very prolonged AE time...

Gints

[koabi]

z0tya:  perhaps it is looking at the values on the row below is the only thing I can see.  I need to get effective VE value to be what is expected in order to move forward.

GinstK, I think 4ms is fine here since it is going to a DIS amplifier.  I do not remember why I set the req fuel too low - I think I was compensating for a problem in a crude way and never went back to correct value.  I do not believe hydro hammering is a problem with this fueling setup but I will monitor fuel pressure more closely.

I wish, if possible, to leave the car with speed density since it will be fitted with a new cam for a future turbocharger.  I do not know what is considered a clean enough MAP signal.

I will try 1.2.0, I will change dTPS curve, and I will try to refactor VE based on a proper req_fuel value.

Any thoughts on what to expect with a decrease on the injector effective rampup value?

[GintsK]

I mean - your dwell goes low at high rpms!

Req fuel can stay different. I just note - volumetric efficiency is still low if you heave just 1200cc.

You can try to solve noisy map with higher count of MAP filter values. And may be rearrange VE table bins more closer at 80....100kpa region.

BTW it is possible to manage Alpha-N with boost on VEMS.

Regarding injector dead time. Documentation for me seems wrong. You can try to understand my findings if "traditional" method is used. Or simply use much more straightforward "simplified" method. http://www.vemssupport.com/forum/index.php/topic,1261.msg13418.html#msg13418

Gints

[koabi]

Gints,

I ended up leaving VE mostly alone and tried the coil map scale change and the car ultimately picked up nearly 1 full second in the 1/4 mile from last week to this week   Another car with the same ignition system and configuration does not have this problem but I think we'll change the other car's settings as well.

We still have fueling to improve and the decision for alphaN but I am pleased with the results so far - thanks for pointing the problems out!

koabi

[GintsK]

Did you make some datalogs?

[koabi]

I reviewed the datalog (http://vems.hu/vemstune/sharingcenter/reports.php?cmd=view&key=TOHTIe) and was surprised to see the dwell time decrease as it had before.  Go watch after 45 seconds into the log to see the run.  The coil charge time map scale configuration change from 0 to 1 seems to have had little effect.

We also replaced the ignition amplifier as a precaution at the same time as changing ignition settings, so I think that was the true cause of the misfire at 6000 RPM and above.

and for fun, here's the video of the best pass of the night:
http://www.youtube.com/watch?v=aLFEIlsr1N8