fbody_brian

Figured I'd ask here. This is a duplicate of a post I made on another forum, maybe someone here has some insight.

I've searched a lot and have found other people with the same issue, but never any solutions.

I get a lot of fluctuations in my rpm at WOT. the logs look really jerky. Like it will go up a few hundred rpm and then drop back down a little then repeat.
Strange thing is I don't feel it at all and the car seems to run strong for what's done to it.

I thought that it was possibly fuel knock as my knock counts were climbing last time I logged at the track when it happened. I removed some timing and installed a 160 stat and got rid of the knock counts. Yesterday I got 0 knock counts but the log is super rough, with the rpms fluctuating the whole WOT run through 2nd and 3rd gears

Things I've done so far.
Changed the plugs, checked fuel pressure, cylinder drop test, tested injectors, logged AFR with wideband.

Haven't done a compression test, it's on the list.

Things that are questionable.
Transmission does strange things sometimes, not wanting to shift from second at moderate part throttle, ALWAYS shifts at full throttle. Also sometimes it will pulsate on decel, like it's grabbing and slipping, but not always. Has never had engine braking when when in manual 1-2-3. Not sure about the torque converter, went with a dacco reman unit because of tight funds.

Rear end, not sure about the condition, has some seeping at the axle tubes, sometimes get a slight clunk when accelerating from a very slow coast. It came from a 02 manual car. don't know the history. I'm not easy on it.

Engine was freshened up by the previous owner. New rings and bearings. Had about 1k miles on it when I purchased.
Found a lot of stuff wrong with the car that was not disclosed by the PO, so everything is questionable.

Anyone know what to look at next?
seen this before?
I'm at a loss here.

Thanks

Injuneer

Knock counts aren't knock retard. Look at the knock retard number - that is the one that matters. Knock counts is a constantly incrementing field, until it gets into the 60,000's and then it resets. Typically you will get a few thousand count increase on cold start. But that doe not necessarily cause knock retard. Seeing that field stay at "0" would be very unusual.

Ever look closely at a dyno graph, when there is little or no "smoothing"? It can look like a saw blade, particularly at the top end. If, however you get to a certain RPM and is loses a lot, then increases steadily to around the same RPM, and drops the amount, could be valve float, for example. The 1.7X RR's may be a problem.

Slipping transmission components can cause the problem. Have you scanned it for codes? There are transmission codes that do not turn on the SES light.

IF you have a data log, post it. What scan system are you using? Post in .csv or Excel format.

fbody_brian

Yeah, I'll look at knock retard when I can get to my laptop this evening. The logging software has a knock graph that plots out knock, I had none. I do get the knock incrementing on start, but I got none during the drive.

Lift with the 1.7s is .510/.521
Max Lift on springs (pac-1212x): .550
It happens starting at around 3500 and does it to redline. It's strange, especially since I don't feel any kind of missing or loss of power.

I think that it did it before the rocker install though, I'll have to dig out an old datamaster log. But I'm really leaning to it not being a spring issue. Could be wrong though.

I'm getting no codes, but the trans is definitely a concern.

Was previously logging with datamaster, but I've switched to eehack exclusively now. It works natively on my preferred OS and can log and flash. I'll post some logs or a csv file later.
Thanks

ACE1252

Here is the software he is using to log with....
New $EE tuning tool! - LS1TECH

fbody_brian

Here's a log from my drive to work. there's a wot section about in the middle.
I've included both the $EEhack log and a csv export. I'll include the whole log and just the wot section of it if it's easier.
I've renamed the .eedata files to .txt, so if you need to open them in eehack you'll need to rename them back to .eedata

Injuneer

Looked at the WOT. There are some significant RPM drops. What does stand out is that the MAP is jumping around a lot. Normally, as RPM increases, MAP will drop off gradually.... that's the frictional losses through the inlet track as flow increases. But you have places where the MAP drops by more than 4 kPa, and then suddenly recovers.

Sticky valve, misfire, hard to tell. But if you add another column next to RPM and calculate RPM delta, and another column next to MAP and calculate MAP delta, there is a loose correlation. Cause/effect to be determined. Which is driving which?

Definitely have a dead left side O2 sensor, which is keeping it out of closed loop, not allowing fuel trims.

Is the time stamp in milliseconds? Appears to be.

fbody_brian

The left O2 is not plugged in, it's currently set to stay in open loop. My wideband is plugged into that header.

I believe that each record is a 10th of a second. Not sure though.

I do think I hear a lifter clacking sometimes at idle, but it's hard to pinpoint, wonder if I have one going bad.

Thanks for taking a look

Injuneer

Does your wideband controller have a narrow band emulator capability? That would supply the 0-1000 mV signal the PCM is looking for.

The delta between records is typically "90" which led me to think 90 mSec, which would be 0.090 sec, or a shade under .1 seconds. I had not looked at the full data log, but when I just did, the time stamp has the decimal place where it would logically be. The WOT file did not have a decimal.

fbody_brian

Yes, it has a narrow band emulator. Just haven't wired it up yet.
I actually just deleted the portions of the log that were outside of the wot section. It
Probably messed up the timestamp when I did that. I was trying to find some way to just save that section. Probably should have exported it to csv first.