[cnstman]

I am experiencing a very weird problem.

Background: Last weekend i had an APS turbo inlet pipe (intake manifold was of course removed for this), JECS injectors, and a td06h-20g turbo installed. I made a base map with scale and latency values as follows: Scale Value: 3680, Injector Latency 4.68 2.38 1.29 0.92 0.67. this map started life as the rescalled map that WolfPlayer created, then cobb converted to the 04 .stb

Problem: From the approx rpm range of 2000 to 2600 the car freaks out. the short term fuel trim maxes at 25.xx, the a/f goes lean, very lean sometimes. and the car stumbles and hesitates. after 2600 to 3000 or so the car runs perfectly. last night we checked for vacuum leaks, and did not find any.

Questions: is there anything that the ecu is doing during this rpm range that would cause this? Would unhooking the TGV durng the install have any ill effects? is there a method to hook the TGV back up, or do they just start working again once reinstalled?

My next step from the mechanical side is to take the manifold back off and see if the gaskets are leaking. My mechanic has 20+ years experience, and 18+ on suby's. but everyone makes mistakes so it is not out of the question that something is leaking and somehow seals itself up as rpm's rise.

My next step from an electrical / Tuning side is to try a new base map and see if the map i am using has a glitch in it that is unseen on the dash board or logs.

the following is a link to 2 logs from last night that catch the event. The DAM is .7 because we had just reset the ecu. before the reset the DAM was 1.0.
http://premium1.uploadit.org/cnstman...-Term-Logs.xls

Thanks for your time, and help.

i should add, i have the stock intake with a K&N filter (cleaned last weekend also), the perrin MAF hose after the stock MAF housing, then of course the APS inlet hose.

[WolfPlayer]

Wow. Those shortterm trims really do look whacky. FWIW, those people who I have talked to with injectors have some whacky trims ... but not that bad. Every now and then I get some 15-20% trims like you. However, they never hold there. Could the turbo inlet pipe be flowing that much air to throw things off by that much? I don't know. Makes me worry about installing my TGV deletes and inlet pipe though. Hmmmm. I would recommend INCREASING your Acceleration Enrichment. Your car goes lean when you crack the throttle blades. Increase values in that table by 10% and then retest. Increase by 10% and restest again. It will take some time but I think this will help you out a lot (so long as it is a tuning problem and not a mechanical problem such as a vacuum leak).

t

[cnstman]

Thanks for your responce. Although the logs show this happening when i get on the gas, it actually hapens every time the car passes through this rpm range, even if i get on the gas at around 1500 and pass through the range with a constant throttle %. i am leaning toward a leak at this point, even though we sprayed every location we could see with carb cleaner to try to find it. we will be removing the intake and installing new gaskets in the next couple of weeks.

still need to flash a new base map to see if there was a problem with an unseen map that could be causing it.

[daemon]

I had a similar spike, though not nearly that extreme in the same area (around MAF=2.5v) a while back. I'm pretty sure mine was a leak that required the turbo to start sucking a bit before it would appear.

[RainMaker]

turbulence in the maf is my guess....and I havent figured out what to do about it other than install some honeycomb

[STirish]

Originally Posted by cnstman

I am experiencing a very weird problem.

Background: Last weekend i had an APS turbo inlet pipe (intake manifold was of course removed for this), JECS injectors, and a td06h-20g turbo installed. I made a base map with scale and latency values as follows: Scale Value: 3680, Injector Latency 4.68 2.38 1.29 0.92 0.67. this map started life as the rescalled map that WolfPlayer created, then cobb converted to the 04 .stb

Problem: From the approx rpm range of 2000 to 2600 the car freaks out. the short term fuel trim maxes at 25.xx, the a/f goes lean, very lean sometimes. and the car stumbles and hesitates. after 2600 to 3000 or so the car runs perfectly. last night we checked for vacuum leaks, and did not find any.

Questions: is there anything that the ecu is doing during this rpm range that would cause this? Would unhooking the TGV durng the install have any ill effects? is there a method to hook the TGV back up, or do they just start working again once reinstalled?

My next step from the mechanical side is to take the manifold back off and see if the gaskets are leaking. My mechanic has 20+ years experience, and 18+ on suby's. but everyone makes mistakes so it is not out of the question that something is leaking and somehow seals itself up as rpm's rise.

My next step from an electrical / Tuning side is to try a new base map and see if the map i am using has a glitch in it that is unseen on the dash board or logs.

the following is a link to 2 logs from last night that catch the event. The DAM is .7 because we had just reset the ecu. before the reset the DAM was 1.0.
http://premium1.uploadit.org/cnstman...-Term-Logs.xls

Thanks for your time, and help.

i should add, i have the stock intake with a K&N filter (cleaned last weekend also), the perrin MAF hose after the stock MAF housing, then of course the APS inlet hose.

Has this issue been resolved? If not have you checked the seals between the fuel rails and TGVs or the other TGV connections? I have very close to the same symptoms as and I just found that the cylinder #1 has a leak in between the TGV and fuel rail. I found it by pressure testing the intake mani and spraying water onto the injectors, TGV connections, etc. It's going to be a PITA to fix. Seems like several people have had the same sort of issues.

[cnstman]

Originally Posted by STirish

Has this issue been resolved? If not have you checked the seals between the fuel rails and TGVs or the other TGV connections? I have very close to the same symptoms as and I just found that the cylinder #1 has a leak in between the TGV and fuel rail. I found it by pressure testing the intake mani and spraying water onto the injectors, TGV connections, etc. It's going to be a PITA to fix. Seems like several people have had the same sort of issues.

i was able to make it run fine by just tuning around the problem. my problems were around the 2.19 to 2.40 or so maf voltage, so i just increased the air flow in those areas so the computer could add more fuel and work around my problem. my and my mechanic's schedule has been a mess for the last 2 months, we are going to take it all back apart and look for any leaks. will update this thread when/if we find anything.

another symptom has arose related to the same problem i think. when using cruise and going down a medium grade hill, the car will "hunt" erratically for 14.7. the car will buck and jump as the a/f goes from 15.xx to sometimes as low as 10:1. it has to be the right situations for it to happen, to steep a grade the cruise will let off the gas completely and my a/f goes completely lean (like it should) to shallow of a grade and it does not do it. i have been able to duplicate it not using the cruise, but it seams to be better produced using it. seams to happen at a certain throttle % or % change. i am wondering if it has anything to do with tip-in, or tip-out (if there is such a thing) i am also leaning towards a leak causing this latest problem, since the engine would be under more vacuum under these conditions.

[STirish]

Originally Posted by cnstman

i was able to make it run fine by just tuning around the problem. my problems were around the 2.19 to 2.40 or so maf voltage, so i just increased the air flow in those areas so the computer could add more fuel and work around my problem. my and my mechanic's schedule has been a mess for the last 2 months, we are going to take it all back apart and look for any leaks. will update this thread when/if we find anything.

another symptom has arose related to the same problem i think. when using cruise and going down a medium grade hill, the car will "hunt" erratically for 14.7. the car will buck and jump as the a/f goes from 15.xx to sometimes as low as 10:1. it has to be the right situations for it to happen, to steep a grade the cruise will let off the gas completely and my a/f goes completely lean (like it should) to shallow of a grade and it does not do it. i have been able to duplicate it not using the cruise, but it seams to be better produced using it. seams to happen at a certain throttle % or % change. i am wondering if it has anything to do with tip-in, or tip-out (if there is such a thing) i am also leaning towards a leak causing this latest problem, since the engine would be under more vacuum under these conditions.

Is this symptom during, what I would say is, almost no load at all? Or just enough throttle to keep the car from shutting off fuel supply (14.7) during deceleration, but the car is still decelerating? If so I have seen the same behavior w/ my JECS where the ARF shoot back ond forth between 10.xx and 16.xx. I have yet to try and log this issue, have you been able to log the occurrence?

[WolfPlayer]

For the record, you will get at least some very slight bucking with 700+ cc injectors. Even my tune that I work on all the time exhibits this. It happens under 2 conditions: 1) going down a hill in hot weather but not in DFCO 2) idle in traffic in HOT weather.

Fact is, during both of those coniditions the ECU is commanding a very low injector pulse width. 700+ cc injectors just can't fire reliably at those commanded pulse widths. The subaru select monitor data stream reports 1.54ms. However, the actual commanded pulse widths are much lower than that. The SSM bottoms out at 1.54ms so we can't see the real commanded value.

As far as 2000-2600rpms, I am betting on the fact that you need more acceleration enrichment. Go out and get a log with your wideband. Do very short increases in throttle and then hold that increase in throttle until you reach a steady state. For example, going down the road at a given accelerator input, give it a LITTLE more input and hold it there. Do this quite a few times with varying changes in TPS. Now review the logs. Mine showed that I always went to 15.x:1 when I did this for just a couple of records. Short terms would immediately increase but then slowly come down as steady state was achieved. I am finding that I have to increase my acceleration enrichment substantially.

t

[LOLSTi]

This is a very informative thread! Thanks Wolfplayer!

[R4ND0M_AX3]

Originally Posted by WolfPlayer

...The subaru select monitor data stream reports 1.54ms. However, the actual commanded pulse widths are much lower than that. The SSM bottoms out at 1.54ms so we can't see the real commanded value...

I'm looking at a log right now that has IPW values of less than 1.54ms.

There's a bunch of 0.77's and a couple of 1.28's. Appears to be during shifts.

[WolfPlayer]

Originally Posted by R4ND0M_AX3

I'm looking at a log right now that has IPW values of less than 1.54ms.

There's a bunch of 0.77's and a couple of 1.28's. Appears to be during shifts.

I'm sorry R4ND0M, I should have clarified further. Normal driving. No DFCO, No shifting, etc. If you find different, let me know. Everyone with bigger injectors gets 1.54ms at idle and very light, no load, cruise conditions. Funny thing is, I can command lower fuel by using lower latency or a lower scalar but the IPW always remains the same. A good example would be you and me. 720cc injectors definitely require less pulse width than the 850cc injectors. We are both at 1.54ms at idle. The helix injectors showed the same.

... maybe it doesn't have to do with the SSM. However, I CAN command less fuel via the latency or scalar and get less fuel ... yet the IPW remains the same in the SSM.

t

[cnstman]

Originally Posted by STirish

Is this symptom during, what I would say is, almost no load at all? Or just enough throttle to keep the car from shutting off fuel supply (14.7) during deceleration, but the car is still decelerating? If so I have seen the same behavior w/ my JECS where the ARF shoot back ond forth between 10.xx and 16.xx. I have yet to try and log this issue, have you been able to log the occurrence?

Finally caught it on a log today, this is in 3rd gear, coming down a hill, around 30-35mph. the throttle changes are cause the car was bucking so hard my entire body was moving causing the throttle to move. a/f is all over the place (m-300 USB) short term fuel trims are also all over. HELPPPPPP!!!!

http://premium1.uploadit.org/cnstman...ucking-log.xls


here is another log, same story, nevermind the long term fuel trims, i was experimenting with something else in this log

http://premium1.uploadit.org/cnstman...iation-log.xls