Maximum VE vs Logged VE

[i_s_64]

Hello,

I'm building out a fuel map on a car, and finding that my VE ranges from 50 in idle (as expected) and as high as 150 in high RPM, high boost

While driving the car around, it's great in idle and cruise, but right around 5000RPM and 7psi of boost, it's like the car has a rev/boost limiter, but it eventually creeps through there to higher numbers up to 18psi of boost. No rev limiters or boost limiters are active, I have been chasing this issue for a few weeks now. I can sit on the driveway without the car in motion and free rev the car without any load and it will go right to the 7000rpm rev limit.

Today I had a look in the logs and noticed my VE has been doing something strange this entire time. Right when I get to 127.5 VE, the next commanded VE will be 0 in the log but the next commanded value in the fuel map would be 128, however the injector duty cycle continues on in a stable manner and continues to increase. 127.5 VE happens at 5000rpm and around 7psi of boost.

Considering the injector pulse width continues on with no issue, part of me feels like this is a non issue, and it's something as simple as the VE being represented as an 8bit value and once it hits 127 it's truly overflowing to a negative number (int8_t). It would seem like my issue might be elsewhere.
However, the reason why I'm having a hard time ignoring it is because it seems to coincide with the difficulty of being able to "push through" 5000rpm at 7psi of boost. It seems like it's missing on multiple cylinders for an extended period of time before it finally gets past it.

So I'm wondering, is it possible that this overflowed VE number is also referenced by other parts of the code, like ignition, etc, which ends up creating this issue?

Thanks,

Alex

[AndreyB]

Yes.

Also, please upload a log with that.

[mck1117]

The overflowed value is not referenced by anything internal to the ECU, it is just a logging issue. Fix whatever is going on causing VE to need to be so high.

[i_s_64]

Yes.

Also, please upload a log with that.

I have uploaded the tune and log.
https://rusefi.com/online/view.php?msq=1605

Around the middle of the log I tried to hold the car at a constant 5000rpm to exacerbate the problem

[i_s_64]

The overflowed value is not referenced by anything internal to the ECU, it is just a logging issue. Fix whatever is going on causing VE to need to be so high.

I am certainly concerned by the VE going that high, but the car runs great except for getting past 5000rpm. The plugs look great and everything. Tomorrow I might look into if the wideband sensor is being funky. Maybe I thermally shocked it and it's no longer showing correct readings. But as far as that goes I'm pretty much stuck

[i_s_64]

The overflowed value is not referenced by anything internal to the ECU, it is just a logging issue. Fix whatever is going on causing VE to need to be so high.

Thanks. Seems like changing it so the VE stays between 0 and 127 didn't change the issue. I lowered the injector flow constant from 650cc, to 550cc, that allowed me to pull between 10 and 20 points out of every VE cell, so my VE percentage always stays within range. Issue still exists at 5000rpm but the logged VE is fine now.

So it's back to the drawing board on this issue

[zed65]

Fix whatever is going on causing VE to need to be so high.

Just for clarity, on boosted application VE-value can be much higher than 127.

[AndreyB]

Just removed the 127 logging limit https://github.com/rusefi/rusefi/commit/2ef6a82e67e4fae4954bb228af3bd3b8f2c71df4