Hello, I recently did an engine upgrade (see sig) and of course it didn't just run perfectly, had to fix a bunch of things done wrong.

Right now I'm out of ideas. What is still happening is the engine feels hesitant. Idle fluctuates a bit too much, at low rpm it hesitates/bogs down, makes for a crappy ride. Starts up fast enough cold, but sometimes cranks too much when starting it warm.

I did pretty much everything LE recommended but I don't completely trust the people who actually did the work.

When given 50%+ throttle it seems like it pulls ok, maybe a bit hesitant, much more power than before the build but still there's definitely something wrong here. It's very uncomfortable to drive "normally".

I attached some logs I took tonight. They progress from cold start, cold running to full temperature running. Run6 I think is the most useful since it is the perfect example of my problems: bogging down a lot while running at 1500-1700 RPM constantly.

What I can easily notice from the logs, but have no idea how to interpret, is:
1 - Fuel trim cell shows 18 at idle and 6-8 while cruising at 1600rpm
2 - STermCounts have a tendency to go down to 110 from time to time
3 - Spark advance jumps up from 40 to 44 and goes down to 35 then goes back to 39-40
Apparently 2 and 3 are related as they seem to happen at about the same time. I can guess whenever the advance wonders the bogging/shaking happens.

If you have any ideas please let me know.

http://www.ls1tech.com/forums/attach...er-ee-logs.zip

 

-idle cell should be 16, and it is at the start of several files. It moves to cell 18 when the throttle voltage starts to increase, and the PCM calculates throttle position greater than 0%. When the PCM sees more than 0% throttle, it figures it isn't idling any more, so it moves to the default "load" cell, which is 18. The PCM sees 0.67V and sets that s 0% throttle. All of a sudden, the voltage changes to 0.80V, and the PCM interprets that as 3.5% throttle - it isn't at idle any more. Were you resting your foot on the throttle? If not, the throttle voltage is varying when it shouldn't be.

Look at the long term corrections, not the short term. In almost all cases (cells) the PCM is adding fuel using the long term counts. Its normal for the shorts terms to vary either side of 128.

Its normal for spark advance to vary, since the PCM is looking up the advanve value based on a table that has RPM on one axis, and MAP on the other. As RPM and MAP change, it will pick a different value for spark advance.

You have knock retard showing up in several logs (but not in 6). The sprak advance shown on DataMaster is the spark advance the PCM has found in the table MINUS the knock retard, so whenever you have knock retard, the timing will vary significantly more than it would without knock retard. Add the advance and the knock retard and you will have the value the PCM is picking from the spark tables. But I would agree, the advance seems to be very erratic... in some places in 6 it goes from 40 - 35 - 45 in three frames. I tried to correlate it to RPM and MAP, and I can't. And there's no knock retard. It would be very unusual for the spark tables to vary by 5-degrees with those tiny changes in RPM and MAP.

It does seem strange..... everthing in the log is failry stable, but the spark advance is dancing all over the place. Are you able to directly correlate the "bog" to the points where the spark advance becomes erratic?

Why do you have DataMaster set to "Y-car" (Corvette)?

 

I can't really say the spark advance variation points are exactly when the car bogs/shakes as I was looking at the road while driving but I can tell you the frequency and number sure match what I feel. The graph variation sure "looks like it feels".

And it does make sense, given that the word "shakes" is better suited than "bogs". I know how it felt when it bogged, it slowed down felt like no fuel or something. Come to think of it, only at really low rpm in 1st gear it feels like bogging, from there on it's more shaking. When standing still and trying to move forward or in reverse if I don't rev it enough it almost stalls sometimes, drops RPM suddenly pretty bad.

Another thing maybe worth mentioning, when I switched from normal fuel to "best" fuel the symptoms definitely happened less often. I can't really say though wether it stayed like that or got worse in time.

I have no idea about the corvette setting Didn't notice it.

L.E. At idle the fuel cell is 16. In run6 I was actually getting ready to take off.

 

If by "normal" fuel you mean low octane fuel, that can account for the knock retard. The effects of driving for an extended period of time with low octane fuel seems to linger, with the PCM basing knock retard on knock counts - calculated conditions that MAY produce knock - rather than listening only to the knock sensor.

 

Here in europe there are only 2 types of fuel, normal and "upgraded". They call them 95 octane and +-100 octane.

I haven't rewritten the PCM since I fixed some stuff and changed the fuel. I did have the battery off though.

I'll give rewriting the PCM a go considering what you said. During the troubleshooting I did re-write it a couple more times and it did make things better after fixing some critical problem. It seems to have some memory beyond battery disconnect. I did run the car with 2 injector wires switched (not my mistake I just found it like that).

Since the rebuild I only ran it a maximum of 500 miles.

 

Europe uses a different octane rating system than the US.

Your numbers are based on the "research octane" (measured under moderate load). US uses the average of "research octane" and "motor octane" (measured under heavy loads), so the octane ratings can not be directly compared. Typically, research octane for a given fuel will be at least 10 points higher than motor octane for the same fuel, making it 5 points higher than the (R+M)/2 used in the US. Motor octane is a more meaningful number for severe, high performance applications.

 

Although I already knew partly what the different octanes were about I must say you've explained it better than anywhere else I've seen it explained

 

What is the best possible way to program the PCM - engine hot or cold, start it right after or wait a while, any break-in procedures to make it work better etc?

 

Reprogrammed the car and ran it a little, clearly there is no difference.

I talked to a friend and we're taking the plugs out tomorrow and checking compression too, he suspects the valves aren't adjusted right...

 

The PCM and the engine are two separate things. Why would the temperature of the engine have anything to do with when you program the PCM. Why would changing the program in the PCM require a "wait" or a "breakin"?

Program it and run it... don't worry about whether the engine is hot or cold.

 

What I meant to say was that maybe the PCM could get better attuned to the engine if the engine was running at optimal temperature during the first moments of running or something. I have no idea how it learns and adapts.

 

When you reprogram your PCM, it will clear all the long term fuel corrections. It will be starting from "scratch". But the PCM can't "learn" until the engine warms up and the PCM enters closed loop. Let the PCM start the engine, cold or hot, and the PCM will do what it needs to develop the long term fuel corrections. It can take a while. The car has to be driven over a wide range of conditions, varying RPM and engine load (MAP) to develop the long term corrections in all 16 cells defined by those operating conditions. Some cells don't get used very often, so it could take a while for them to stabilize.

But again, don't worry about it. Its no different that starting a car with a stock tune, after pulling the power to the PCM and clearing the learned values.

 

We took out the plugs, took pressure readings and adjusted the valves.
- 1 plug was completely black, 2 were half black
- 1 cyl wasn't making much pressure, 2 were slow to do so
- valves were of two kinds: some didn't close and others didn't let the lifters fill up.
Ran like this for 300 miles or something. No point explaining why this happened. Sufficient to say it's hard to get decent help around here.

Fixes:
- re-checked gap and shuffled plugs back in
- released all valves until zero lash (2 complete turns) and ran the engine (5 minutes) until all lifters started push oil up the pushrods on the rockers;
- with all lifters solid we went zero-lash + small preload, started engine, no clicking/noise.

Results:
- smoother and more stable idle, better revving
- faster starting up
- exhaust still smells like fuel, brings tears to eyes and chokes
- bogging/shaking still there, but significantly less violent - still not good enough to let pass.

I'll take some more running datalogs tomorrow and see how things show.
Here is a log recorded at idle after we finished today.

LOG TAKEN AFTER FIXES

 

Did you clear the PCM before restarting, after setting the valves?

 

We reset it by disconnecting the battery.
We had the laptop connected and watched it go back to almost exactly the same LTermCounts as soon as it entered closed loop. Pretty strange how doing the valves and making all cylinders work had no effect on LTermCounts. Especially the passenger side 160 value...