Lower VE with newer/better engine?

[JakeRobb]

I'm a newbie at this, so I've probably made some incorrect assumptions here. Please set me straight!

I'm working on tuning my new engine. It's an LS3 with ported heads and a custom cam. I've also added headers and a larger intake tract. The LS3 has a 24x reluctor and all the appropriate LS1 sensors, and I'm running it with the stock P01/0411 PCM from my '02 LS1 F-body.

I am just using my narrowband O2 sensors so far; still need to wire in the wideband (planning on wiring it into the EGR circuit so I can datalog it directly from the PCM). I don't expect to be able to dial in a perfect tune like this, but I'm thinking I can get close-ish.

Initially, when I first had all the hardware hooked up, I started the engine on the completely-stock LS1 tune. It sounded like a normal cam+headers setup and idled fine; I only wanted to confirm fuel and oil pressure, and then shut it down. I didn't datalog this at all, but I assume it was running very rich since the it was driving 42# injectors with 24# injector data. Here's a quick video.

I then took my factory LS1 tune, updated the displacement, supplied MAF and injector data, and applied the RTT OS, then flashed that to the PCM, leaving the LS1's factory VE/PE/idle/etc tables as a starting point.

The engine won't even idle; I have to hold the throttle open a bit when starting, and then continue to hold it there for it to run continuously. Even after doing that for several minutes so that it's warmed up and I can start to see fuel trim values in VCM Scanner, it dies immediately when I let off the pedal. I assume this is an IAC issue due to the upgraded throttle body, but haven't looked into that yet because I was advised to tune VE before idle. This situation has me thinking I should probably do idle tuning first. Or, maybe: does the PCM need IAC calibration data like it does for the MAF? Another possibility: I went with a relatively inexpensive Speed Engineering 92mm throttle body before learning that most tuners won't even bother with those and insist that people use one of the more expensive TBs. Could this refusal to idle be a manifestation of my cheap TB? (I'm prepared to replace it; just hadn't heard about the cheap ones causing this specific problem.)

But, I digress; that's not what I came here to ask:

I had assumed that my stock VE numbers would be close-ish to right, and that for the most part I would be increasing some cells, primarily in the mid-RPM range, to account for the fact that the modified LS3 can flow so much more air.

However, on first start, I'm seeing O2 sensor data indicating a very rich condition. Steady ~0.9v from all four narrowband O2 sensors, with no fluctuation shown on the graph, plus a strong unburned fuel smell. If I understand VE tuning correctly, then a rich condition means that the VE number for the relevant cell(s) is too high. Do I have this right?

Running with that understanding, I focused on the 4-6 cells where I could hold the RPM steady just above idle. 1200-1600rpm, 40-50kPa. Using the RTT VE table, I slowly reduced the numbers from mid 60s all the way down to mid 30s, at which point I started seeing the expected fluctuating signal from the narrowbands -- cycling quickly and steadily between 0.2v and 0.8v.

I'll obviously confirm all of this once I have the wideband hooked up, but it's going to be a while before I have time for that, and I'm hoping to refine my understanding of the situation in the meantime.

So, to get out of that rich condition, I took out what feels like a TON of fuel. It feels weird to me that the volumetric efficiency of my new engine would be lower than my old one -- especially by so much! Is this a valid feeling?

It's possible I have the wrong MAF or injector calibration data. I'm 98% sure I got the injector data right; less sure about the MAF. Is there a definitive way to test this stuff for correctness?

I used OEM LS3 injectors, and copied the data from a 2014 Camaro SS tune (had to translate from gen4->gen3). The MAF is the 100mm from Hawks, which purportedly uses the "OEM GM LS3/LS7 MAF SENSOR," so I got the MAF calibrations from the same tune as the injectors.

I imagine it's also possible that the throttle body is a factor here, but I can't think of how.

I'm sorry, this is long. I'll stop rambling now. Thanks for reading! Any/all advice and feedback is much appreciated!

[ns158sl]

A tune file and a log would maybe help? Also is a Hawks a good tb? i like factory. Also also, what is a MAF calibration?

[SiriusC1024]

You'll still have to tune the MAF. Can't just bring in the cal for another car.

Post the tune and log. Supply injector part number.

[JakeRobb]

A tune file and a log would maybe help?

I'll happily post a tune and log file later; for now I'm just looking to understand a couple things. Does it make sense that my higher-flowing engine would have dramatically lower VE?

Also is a Hawks a good tb? i like factory.

The factory LS1 TB would not have been up to the task of feeding the new engine.

The Hawks product I'm using is a 100mm MAF sensor body. The actual sensor within is an OEM LS3/LS7 MAF.

Also also, what is a MAF calibration?

VCM Editor -> Engine -> Airflow -> General -> MAF Calibration -> Airflow vs Frequency

It's a mapping of frequencies signaled by the MAF sensor and the actual airflow corresponding to each of those values.

[JakeRobb]

You'll still have to tune the MAF. Can't just bring in the cal for another car.

How is this done? (I mean how do I determine the correct values, not how do I apply the values to the tune).

To be clear, it's an OEM GM MAF sensor, and I brought in the factory calibration for another car that uses that exact same sensor. Although it's in a different housing (larger diameter), so I guess it makes sense that a given reading from the sensor would correspond to a higher actual airflow.

[SiriusC1024]

https://forum.hptuners.com/showthrea...-the-Same-Time

[JakeRobb]

... that thread seems to be a huge disagreement on the proper approach. I find myself siding with the guy who says you can't solve for two unknowns given only one formula, and having no idea how to follow the other guy's advice.

They talk about the Smokeshow method, and there's a link to it. It sounds like that makes a much more logically snd mathematically sound approach. Any reason I should not use that method?

[JakeRobb]

Also, that thread seems to be for gen4 PCMs. I'm tuning a gen3 (P01/0411).

[JakeRobb]

Hptuners

Found this thread, which seems to explain a straightforward approach. To summarize:
1. Disable the MAF; tune pure VE like you would a speed-density tune.
2. Re-enable the MAF, then start tweaking the MAF calibration until the tune is right again.

[SiriusC1024]

The DynAir method works really well, actually. It also applies to Gen 3 if Volumetric Efficiency Airflow can be logged.

But if you prefer classic that's fine, too.

Set the tune up first:
Airflow>Dynamic>High RPM Disable to 200
Fuel>Cutoff DFCO>Enable ECT 248
Fuel>Power Enrich>Min Map 85kPa and >Throttle>Hot/Cold to 85%

Then log. Make corrections based on LTFT and STFT plotted vs MAF Frequency. Copy the column values from the MAF Airflow vs Frequency table in the tune to the histograms in scanner.
MAF histo.png

Copy the MAF histo, paste special %-half into the tune table. Smooth by hand. Interpolate/Extrapolate missing data points. Do a final smooth with the tool.

Repeat until fuel trims are within 3%.

I should mention that if you're using LTFT, you need to clear them with Scanner after uploading a tune revision.

[edcmat-l1]

Does it make sense that my higher-flowing engine would have dramatically lower VE?

The answer to that question specifically, is at idle and very low throttle angles, yes. A larger camshaft reduces pumping efficiency at low engine speed.

[edcmat-l1]

How is this done? (I mean how do I determine the correct values, not how do I apply the values to the tune).

To be clear, it's an OEM GM MAF sensor, and I brought in the factory calibration for another car that uses that exact same sensor. Although it's in a different housing (larger diameter), so I guess it makes sense that a given reading from the sensor would correspond to a higher actual airflow.

Any time you change any THING that pertains to a MAF sensor, that includes placement, size of the tube etc., you completely change the calibration.

[edcmat-l1]

I'm a newbie at this, so I've probably made some incorrect assumptions here. Please set me straight!

I used OEM LS3 injectors, and copied the data from a 2014 Camaro SS tune (had to translate from gen4->gen3).

That's probably not a good idea if you're not experienced. There's some conversion spread sheets floating around here you might want to search out. If you tried converting the data yourself and you're wrong, you'll be baking in a bunch of bad data.

[SiriusC1024]

Yep good point. Let's see the tune.

[JakeRobb]

The DynAir method works really well, actually. It also applies to Gen 3 if Volumetric Efficiency Airflow can be logged.

I can read this two ways:
1. some gen3 setups can do the DynAir method; you need one that can log VE airflow
2. you should be able to do this on gen3, assuming you can log VE airflow on gen3 -- but you don't personally have experience with this on gen3.

I think you mean it the second way.

In either case, I know Scanner on my setup shows an airflow value in lbs/min which I can log. Is that what we're talking about here?

But if you prefer classic that's fine, too.

Is "classic" the method I mentioned in post #9? I don't know what I prefer; I have no experience either way. Are there advantages/disadvantages between the two?

Set the tune up first:
Airflow>Dynamic>High RPM Disable to 200
Fuel>Cutoff DFCO>Enable ECT 248
Fuel>Power Enrich>Min Map 85kPa and >Throttle>Hot/Cold to 85%

I think these are the same "disable the MAF" settings recommended for the "classic" approach. I've seen a few variations on how exactly to go about disabling; some people set the high-RPM dynamic airflow cutoff to 0, or 500 -- but that value is not really important as long as it's below idle RPM. Right?

Then log. Make corrections based on LTFT and STFT plotted vs MAF Frequency. Copy the column values from the MAF Airflow vs Frequency table in the tune to the histograms in scanner.
MAF histo.png

Copy the MAF histo, paste special %-half into the tune table. Smooth by hand. Interpolate/Extrapolate missing data points. Do a final smooth with the tool.

Repeat until fuel trims are within 3%.

I should mention that if you're using LTFT, you need to clear them with Scanner after uploading a tune revision.

I think this all makes sense -- but you're describing some pretty fancy stuff in Scanner that I wasn't aware it could do! I think I'm following what you're describing, but I've never seen that Graph Display Layout Editor from your screenshot. Will have to study that bit.

Also: in this process, when do I make my VE changes? I'm assuming before, with MAF disabled -- because otherwise the trims going into the MAF cal would be affected by any inaccuracies in the VE table. Does that sound right?

[JakeRobb]

The answer to that question specifically, is at idle and very low throttle angles, yes. A larger camshaft reduces pumping efficiency at low engine speed.

Any time you change any THING that pertains to a MAF sensor, that includes placement, size of the tube etc., you completely change the calibration.

Ah, makes sense! Thanks for clarifying these. So, unlike injectors where the manufacturer/vendor can supply data, that's not really feasible for a MAF because it's affected by all the other variables.

My cam is a custom grind from Katech -- they took the specs for their relatively old LS3 Torquer cam, and built it with a more modern lobe profile. Specs are 226/239 605/621 113+4. Based on your experience, can you predict where (in terms of RPM and/or MAP) it would start to show an advantage over the stock '02 LS1? It would be nice to have some intuitive sense of whether the changes I find myself making are reasonable.

That's probably not a good idea if you're not experienced. There's some conversion spread sheets floating around here you might want to search out. If you tried converting the data yourself and you're wrong, you'll be baking in a bunch of bad data.

I copied the data from someone who used the spreadsheet and posted their results on these forums. It was months ago now; I don't have details. Still, I'm reasonably confident I have it right.

EDIT: I got my data from here.

[SiriusC1024]

All those settings other than High RPM disable will be used to tune either MAF or VE. WHen you do VE the only difference is Engine Diag>Fail High to 0Hz, P0101-P0103 to MIL on First Error, and disable the Line Presure MAF checks at Trans Diag.

Don't copy the data from another tune without verifying. Like edcmat suggested let's see the tune and verify injector data first. There's Min Fuel in Transients, for example.

[JakeRobb]

disable the Line Presure MAF checks at Trans Diag.

I have a manual trans, so I assume this is not applicable.

[JakeRobb]

All those settings other than High RPM disable will be used to tune either MAF or VE.

Okay -- but VE first, right?

[SiriusC1024]

No, first the tune Don't have to do VE before MAF, but yes that's typical.