If I plot the transient throttle percentage enrichment on that log I posted, it stays at 0% the whole time. If you look at the timescale on that log you will see that I am VERY slowly increasing the throttle to maintain the target vacuum. The delta throttle is almost nothing. Transient throttle should not be a factor here.

If I hold it at 2400 rpm yes it stays lean...given high enough vacuum (down a hill, flat ground at lower speeds/gears). In fact if I free rev the engine to 2400 rpm and hold it it will go off the scale lean. I have tried tuning the far left column (no load) and it is impossible to get cosistent A/F ratios.

The issue is that the lean spots are so narrow (rpm wise) that if I tried to fix the problem by adjusting the fuel curve it would be extremely rough. That area of the map is generally in the 90's for VE. During the lean spot I need a VE value of like 130 to get enough fuel. So I would essentially have a ~300 rpm band that goes from 90 to 130 and then back down to 90. You can imagine how the fuel curve would look if I did that! In no way is it "normal" to have a giant spike in the VE table like that. The bottom line is that there is some other problem here, and trying to tune around it is not the right thing to do. If you try to tune around mechanical problems you will always be chasing your tail.

 

it happens only at 30XX rpm between -8 to -2.5

as you can see in the log my map is so smooth the only problem is the lean spot at 30xx rpm. I tried to put the intake and exhaust timing back to 0 but nothing changed.

could it be the intake manifold is leaking from some where ?
because it only happens in vacuum.

 

very interested in this thread. Too many ppl think haltech is the answer to everything! But the more i read, the more problems i find.

 

Ok for the third or fourth time: THE HALTECH IS NOT THE PROBLEM.

This happened with the stock ECU/UTEC. It is STILL happening with the Haltech. How does that make the Haltech the problem? Read the thread.

 

I did read that part, but it was interesting how you changed to Haltech assuming it would fix the problem.

I've also read early model zeds having issues with the haltech with the upgraded software.

 

I switched to the Haltech with the hope that the greater control would allow me to tune out the problem. Since then it has become clear to me that it is not a tuning problem but instead some kind of mechainical problem. Switching to a different computer is never going to cure a mechanical problem.

That said, I am still glad that I have a Haltech and it is far better than what I was doing before. I have had no firmware problems whatsoever. I just want to make it clear that this is not a Haltech bashing thread, I think the Haltech is a great piece of equipment and I am very happy with it.

 

I'd put a volt meter on your injector wiring. Make sure the voltage is constant, use the engine as your ground reference...this would also rule out engine grounding issues if the voltage is steady.

Also, try putting the volt meter in AC mode and make sure there isn't any AC noise at that spot either (should read near zero). Sometimes DC looks steady but when you put it on AC you'll see all kinds of nonsense.

Is there any way to completely disable the cam phasing on the haltech temporarily? I can't think of anything other than electrical or cam phasing that could be causing this given everything you've already checked (fuel pressure, ect)

 

I tried setting the entire intake cam table to zero, and it did not have any noticeable effect on the lean condition. So I don't think it is the cam phasing. Then again I didn't datalog when I tested so I guess I can't say for sure that the intake cam was really at zero advance like it was supposed to be.

Voltmeter on the injector wiring is the next task. I also sent Hal that datalog so hopefully he can take a look at it.

 

You may want to try the Haltech v1.0 firmware and see if it rectifies the issue. I'm not thinking there is something wrong in the current firmware, but some setting in the original firmware may be covering for an issue specific to 2003 cars (injector/ignition driver/wiring etc).

The + voltage line feeding the coil packs also feeds the intake cam phasing solenoids - and str8dum1's car would stumble with misfires and has this weird cam phasing issue that goes away with v1.0 of the firmware. Maybe there's some odd link - logging that voltage might be worthwhile.

 

sub'd, good info in this thread.

 

Did the v1.0 firmware fix both of those issues? I thought it fixed his misfire issue but he was still having problems with the cam phasing...

 

ya misfire seemed to go away, but no dice on the phasing. Blackbird's does the same thing with his phasing. That makes 2 2003 G35 cars over 600whp that have phasing problems. Not sure if Haltech has any solution as it doesnt do that on any other year (or at least people havent noticed)

 

str8dum1, the v1.0 did not fix the cam phasing issue? wasn't sure it had been tried while on the dyno - nm then I guess. so weird... I have a Ken Bell Boost-a-spark that is easy to splice in if you want to try it. It does nothing for me (on the dyno).

 

I have seen this thread pop up numerous times with no resolution. Good luck.

 

hmm never tried 1.0 on the dyno. probably go for a retune in about 3 months though and will investigate further

 

One thing I forgot to mention - I keep getting the following CEL:

P0444 - EVAP CANISTER PURGE VOLUME CONTROL SOLENOID
VALVE

"An excessively low voltage signal is sent
to ECM through the valve"

This came on after I installed the Haltech. I checked the valve and it is connected and functioning properly. There are a few more tests in the FSM I need to do though.

 

has there been any reports of this occurring on nearly stock non-turbo cars running stock ecu ?

 

Don't think so. I'm pretty sure it's a result of something I did. Other thoughts:

I have tapped into the fuse box for a couple different accessories - namely boost controller and gauges. I know people have had weird throttle control issues as a result of aftermarket tail lights. If tail lights can cause problems, maybe my fuse taps can too?

I did some welding on the exhaust while it was still on the car. I disconnected the battery and put my ground strap on the car during the welding, but I am wondering If I fried something?

 

I have an update:

Tonight I spliced in a hose between my fuel pressure gauge and my fuel pressure regulator so I could stick the gauge under the windshield wiper and watch it while driving (poor man's fuel pressure gauge LOL). Anyway, while driving I was able to confirm that the fuel pressure does not change as the engine passes through the lean spot. It changes with manifold vacuum, as it should, but there is no "dip" like I was hoping to see during the lean condition. So that rules out anything having to do with the fuel pump or return system.

I also thought that maybe all of the accessories that I spliced into the fuse box might have something to do with it...who knows, maybe a ground loop or a voltage drop in the injector circuit. So as a quick test I disconnected all of my aftermarket stuff by pulling the fuse taps - no more wideband, gauges, boost controller, etc. I turned off the O2 control and carefully went for a test drive. Although I no longer had an A/F display, I could still feel the car bog as it went trough the lean spot. Finally, I decided to hardwire the accessories right to the battery instead of the fuse box and once again it made no difference, I still see the A/F go full lean at 2400 rpm. Just out of curiosity, where does everyone get power and ground for their gauges/boost controllers etc.?

So if the airflow is "steady" according to both the MAF and the MAP, and the fuel pressure does not change, and the lean condition happens in both banks, what does that leave? The only thing I can think of is that the voltage in the injector circuit is dipping and causing a momentary increase in injector latency. According to the Haltech, the voltage at the ecu remains constant so it would have to be isolated to whatever circuit is driving the injectors. I guess it's time to start digging through the electrical diagrams in the FSM and figure out where I can tap in and monitor the voltage that the injectors are seeing. I could always pierce one of the wires right near the injector but they are really hard to get to without removing the plenum.

 

Why 2400 rpms? So weird. If voltage drops, your dead time correction should account for it instantly.

Have you taken off the plenum to see if anything is stuck on top of the intake valves? Do you run a spacer? Sometimes those plastic pieces fall off...

I know you tried adding rows around 2400, but have you tried adding columns? I have a col for every 1/2 psi in the mid to high vacuum range going to 1/2 psi beyond what the car ever generates (in terms of vacuum).