dadasracecar
Member
So Ive had this weird semi-violent stutter around 2300rpm in 5th and 6th since owning the car. I started really noticing it after installing the standback on the stock car (boost set at 16 psi). I was just noticing it more though as I was hyperobservant. Several experiments putting it back to stock with the bypass plug indicated that the stutter was there. The stutter to which Im referring was under partial or WOT. Ill be accelerating to merge onto the highway in 5th or 6th from about 2K rpm and as the rpms climb to about 2300 or so theres a sharp hesitation and then it continues to accelerate. You can hear the bov blowing off. Otherwise the car drives normally.
I mentioned this long ago to Jordan and he suggested I datalog it. I could never capture the event as it was very irregular and didnt occur all that often. After the DP install however I have been noticing it with much more consistency. I can generally make the stutter happen at will now. I upped the boost to 17 after the DP and thought that may have had something to do with it but dropping it back to 16 I could still make it happen this morning.
I have two theories about the event. 1st is that the camdriven high pressure fuel pump is unable to deliver the required fuel for 17 psi at 2280 rpm and Im getting a lean spot-detonation-fuel cut situation. 2nd is that the ecu was programmed using some sort of graphical interface that modulated multiple variables to achieve the desire power curve and this is just a glitch.
I datalogged the event finally one morning actually caught it five times. Here are the plots. If you want the raw data to look at you can pm me your email address and Ill send the excel file. Dont ask about plotting vs RPM b/c the RPM fluctuates and would cause the curves to loop over themselves making it even more difficult to read.
Looking at the actual data it seems that there are significant drops in the MAF voltage which is how I picked the rpm where the even took place. Time= 951.3, RPM=2291; Time= 974.7, RPM=2291; Time= 1002.6, RPM=2284; Time= 1033.7, RPM=2226; Time= 1093.7, RPM=2275. The RPM value immediately before each of the rpm values above are 2274, 2278, 2275, 2284, and 2263. You can see that the event is very repeatable. The MAF voltage drops sharply at each of these time/rpm points. The drops are 0.11, 0.86, 1.01, 1.00, and 0.47 V, respectively. Immediately following the voltage pops right back to something similar to what it was before and is after. The injector duty cycle and pulse width both go down at the corresponding points which makes me wonder about theory 1. I would expect that if it was running out of fuel, the PW would go up. That doesnt happen. The TPS is pretty stable to me and doesnt look to be a player in whole thing. The ignition timing goes way up immediately following the event which seems like the lean burn cruising condition of direct injection engines.
Going on theory 1, I added a point to the fuel tables at this rpm from 2.4 to 3.3 V on the TPS and took out a degree of timing at the same places. I was still getting the stutter so I added two points to the fuel. Still there. I think its a little more muted but its hard to say for sure. The fuel didnt seem to do much but Im mostly in closed loop so I wouldnt think it would do that much though I have my doubts about whether this car ever goes into to open loop. Id appreciate any other hypotheses that anyone might have. Its a pretty annoying condition. Here are the datalogs. These are closeups of the time slices where the stutter occurs. PM me if you'd like the data. I'd really appreciate some help on this.
I mentioned this long ago to Jordan and he suggested I datalog it. I could never capture the event as it was very irregular and didnt occur all that often. After the DP install however I have been noticing it with much more consistency. I can generally make the stutter happen at will now. I upped the boost to 17 after the DP and thought that may have had something to do with it but dropping it back to 16 I could still make it happen this morning.
I have two theories about the event. 1st is that the camdriven high pressure fuel pump is unable to deliver the required fuel for 17 psi at 2280 rpm and Im getting a lean spot-detonation-fuel cut situation. 2nd is that the ecu was programmed using some sort of graphical interface that modulated multiple variables to achieve the desire power curve and this is just a glitch.
I datalogged the event finally one morning actually caught it five times. Here are the plots. If you want the raw data to look at you can pm me your email address and Ill send the excel file. Dont ask about plotting vs RPM b/c the RPM fluctuates and would cause the curves to loop over themselves making it even more difficult to read.
Looking at the actual data it seems that there are significant drops in the MAF voltage which is how I picked the rpm where the even took place. Time= 951.3, RPM=2291; Time= 974.7, RPM=2291; Time= 1002.6, RPM=2284; Time= 1033.7, RPM=2226; Time= 1093.7, RPM=2275. The RPM value immediately before each of the rpm values above are 2274, 2278, 2275, 2284, and 2263. You can see that the event is very repeatable. The MAF voltage drops sharply at each of these time/rpm points. The drops are 0.11, 0.86, 1.01, 1.00, and 0.47 V, respectively. Immediately following the voltage pops right back to something similar to what it was before and is after. The injector duty cycle and pulse width both go down at the corresponding points which makes me wonder about theory 1. I would expect that if it was running out of fuel, the PW would go up. That doesnt happen. The TPS is pretty stable to me and doesnt look to be a player in whole thing. The ignition timing goes way up immediately following the event which seems like the lean burn cruising condition of direct injection engines.
Going on theory 1, I added a point to the fuel tables at this rpm from 2.4 to 3.3 V on the TPS and took out a degree of timing at the same places. I was still getting the stutter so I added two points to the fuel. Still there. I think its a little more muted but its hard to say for sure. The fuel didnt seem to do much but Im mostly in closed loop so I wouldnt think it would do that much though I have my doubts about whether this car ever goes into to open loop. Id appreciate any other hypotheses that anyone might have. Its a pretty annoying condition. Here are the datalogs. These are closeups of the time slices where the stutter occurs. PM me if you'd like the data. I'd really appreciate some help on this.
