Odd values with RAFIG for B4307 base airflow

[pinstripebob]

Hi all, I recently installed a camshaft in my 2004 Pontiac GTO. it is Summit Racing part number SUM-8700. I am going through the tuning process, and I seem to be having a weird thing happen with the base idle airflow (B4307) that my engine likes. I have the warm values about 0.5 g/s above the stock tune's values, and if I perform RAFIG it wants me to drop them about 1.3-1.8 g/s. If I log IACDES_B, it gives me about the same number RAFIG would, 6.4 g/s. This seems super low, especially considering the stock values are 7.1 g/s. Modifications are as follows:

2004 Pontiac GTO
LS1 5.7 l
6 speed manual transmission
Stock cable throttle body with IAC
Short headers, JBA mid-pipes with cats, JBA cat-back exhaust
XAIR cold air intake, MAF is enabled
SUM-8700 camshaft, duration at .050" is 225 intake 236 exhaust

I have attached my tune for reference. The VE table is still a bit rough, but AFR is right where I want it to be and I am still doing some logging/tweaking in this area via CALC_VET. Any thoughts would be greatly appreciated, and I can get a log without issue if there are certain PIDs that would help diagnose what's going on.

A few symptoms I'm having that I believe are related to the airflow are as follows:
- Bucking at cruise (45-55 mph) when off throttle
- Bucking at cruise occasionally with slight throttle (TPS at 5.1% or less)
- Engine speed drops to about 450 RPM and comes back up to about 800 RPM when cruising and the car is put in neutral, this sometimes cycles if I continue to roll in neutral at about 45-55 mph
- Engine is hard to start when warm and stalled, although this may be related to B4343 being too low potentially

Thanks!

22434

[joecar]

Please post stock tune also.

[pinstripebob]

Not a problem! I have attached the factory stock tune, as well as my latest tune prior to installing the cam (December 16, 2016).

22439
22440

[statesman]

A few symptoms I'm having that I believe are related to the airflow are as follows:
- Bucking at cruise (45-55 mph) when off throttle
- Bucking at cruise occasionally with slight throttle (TPS at 5.1% or less)

Bucking is rarely an airflow issue. You can try adding some air with your throttle cracker, but I don't think that adding air will help at all. Your bucking is most likely the result of a combination of running lean and a poor spark strategy. Your MAF curve has been beaten up pretty bad and it's low in most places where you would be experiencing bucking.

Just looking at your tune file, I can see you've got a wickedly rich decel. Fix that first. Set all values in {B4003} and {B4004} to 0.75 and then redo your calc.VET. When you've fixed the rich decel, your MAF should go back to something similar to a stock curve when you do the calc.VET.

[pinstripebob]

I'm not super worried about the bucking at the moment, I'm more interested to know why my RAFIG values are trying to push my B4307 values so low.

I still have the stock injectors, aren't B4003 and B4004 tied to the injectors' mechanical performance? Looking at some of my logs, my deceleration has normally been a little lean already. I've attached a log, and if you look at frames 3894 through 4011, you'll see an area where I'm coasting in gear and off the throttle. There is some bucking, characterized by the varying RPM. My wideband AFR averages 14.90 during these frames.

Regarding the MAF table, at this point in the log my MAF frequency is between 3282 and 3403 Hz. Looking at my tunes before and after the cam, this portion of B5001 is within 2.8% of the stock values. The biggest changes to B5001 have been in the range of 8250-10,875 Hz. I might not be seeing the same thing you are, would you mind elaborating a bit more so I can understand what you're looking at?

22445

[statesman]

I'm not super worried about the bucking at the moment, I'm more interested to know why my RAFIG values are trying to push my B4307 values so low.

RAFIG values are so low because your airflows don't line up. Desired airflow has to line up with actual airflow for RAFIG to work.


I still have the stock injectors, aren't B4003 and B4004 tied to the injectors' mechanical performance? Looking at some of my logs, my deceleration has normally been a little lean already.

Okay, you're not pulling enough vacuum in your intake manifold for the decel to go rich. Decel from higher RPMs might still cause a rich decel... if that happens, then you'll know which tables to look at.


Regarding the MAF table, at this point in the log my MAF frequency is between 3282 and 3403 Hz. Looking at my tunes before and after the cam, this portion of B5001 is within 2.8% of the stock values. The biggest changes to B5001 have been in the range of 8250-10,875 Hz. I might not be seeing the same thing you are, would you mind elaborating a bit more so I can understand what you're looking at?

Looking at your log, your fueling looks reasonably good... you've got a rich tip-in and a lean tip-out, but that ain't anything to worry about. Being stock injectors with GM data, you have to expect anything to happen to the MAF curve... so even though I think the curve looks a bit low and messy... that's what it has to be.

I've made a couple of changes to your tune file... see if this helps with the bucking and see if RAFIG works any better.

[pinstripebob]

I actually figured out the B4403 and B4307 correlation last week, I was hoping to get things nailed down a little more solidly before posting my results. I found posts #46-#48 in the thread below and that helped me understand why adjusting B4403 with a stock throttle body makes sense. I have cracked open my throttle body more than stock since adding the camshaft to get my IAC count in the 50-60 range at idle, but had not adjusted B4403. I shifted it 5 cells (it looks like you shifted it 3 cells in your tune) and so far my DYNAIR and IACDES_B have been extremely close at idle and operating temperature.

Here's that thread I mentioned above: https://forum.efilive.com/showthread.php?14439-Idle-Tips-amp-Tricks&p=193144&viewfull=1#post193144

So my B4403 and B4307 are in a very good spot based on my logs. I went ahead and ran through your version of my tune, Statesman. I think I understand everything you did, but correct me if I'm wrong.

1. Modified CL mode so it is in a lower CL mode at lower airflow. Likely to smooth out the transition from low to high airflow/CL mode.
2. Slowed down the rate at which STFT are adjusted, and made them adjust less aggressively at lower engine airflow values.
3. Gave my ECU more room to adjust the airflow correction.

My only question would be with B4401. I haven't seen anyone adjust this before, and my steps are normally well below 200 let alone 310. Although now that I write that, looking at my RAFIG log from yesterday (attached below), my car starts up with just over 200 steps. In that log around from 12,000 I stopped at a store for a minute before finishing my drive home. Restart had IAC steps at 136, nothing crazy.

Anyway, I have attached my tune incorporating your modifications (although I kept my B4307 and B4403), as well as a CALC.VET log and a RAFIG log. Please note that the RAFIG log is my drive home from work, go to frame 19,000 for some steady state, operating temperature idling. The CALC.VET log looks alright to me, although it's showing about a 5% discrepancy in my MAF table. I'm not sure if this is due to the much colder temperature (40s versus 90s from previous logs, Fahrenheit), or if things are all of a sudden that far off. I did an oil change this last weekend, but I doubt that is affecting my tune.

My RAFIG values seem to be much more realistic, bucking while driving has decreased drastically, and the engine at cold startup seems to run much more smoothly. I did notice some bucking while pulling away from a stop when I let out the clutch, I'll need to do a proper RAFIG log and update my B5001/B0101 tables.

Tune: 22468
CALC.VET Log: 22467
RAFIG Log (driving): 22469

[statesman]

I have cracked open my throttle body more than stock since adding the camshaft to get my IAC count in the 50-60 range at idle, but had not adjusted B4403. I shifted it 5 cells (it looks like you shifted it 3 cells in your tune) and so far my DYNAIR and IACDES_B have been extremely close at idle and operating temperature.

Yeah, the adjustment I made was just to get it more 'reasonable'... it was never going to be exactly right. If shifting it 5 cells gets the air to line up properly, then that's where it needs to be.



My only question would be with B4401. I haven't seen anyone adjust this before, and my steps are normally well below 200 let alone 310. Although now that I write that, looking at my RAFIG log from yesterday (attached below), my car starts up with just over 200 steps. In that log around from 12,000 I stopped at a store for a minute before finishing my drive home. Restart had IAC steps at 136, nothing crazy.

I've set it to what it should be. I don't usually explain why I make changes... but when I make changes, they do have a reason. :grin:




The CALC.VET log looks alright to me, although it's showing about a 5% discrepancy in my MAF table. I'm not sure if this is due to the much colder temperature (40s versus 90s from previous logs, Fahrenheit), or if things are all of a sudden that far off. I did an oil change this last weekend, but I doubt that is affecting my tune.

When I saw your MAF curve, I knew you had to be lean. :laugh:

The MAF sensor is immune to temperature changes. I think that having those aggressive short term trim corrections may have masked the lean condition.




My RAFIG values seem to be much more realistic, bucking while driving has decreased drastically, and the engine at cold startup seems to run much more smoothly. I did notice some bucking while pulling away from a stop when I let out the clutch, I'll need to do a proper RAFIG log and update my B5001/B0101 tables.

Your bucking when pulling away from a stop is probably a lean condition. You've still got a lot of tuning to do... but at least you've got your airflows lined up and your bucking is reduced, so it should be a bit nicer to drive now.

Your MAF curve doesn't look right to me. I've made a few more 'adjustments' to your tune file. Do a log on a fully warm engine and log your short term fuel trims and 'GM.VOLTS' as well as all the stuff you've already been logging. I don't need or want any WOT pulls... I'm only interested in your closed loop so avoid power enrichment.

22470

[joecar]

Also, you might want to clean your MAF, just in case (use only CRC MAF cleaner).

[pinstripebob]

Good call Joe, I have some CRC MAF cleaner, I'll give it a spray to be safe.

Statesman, do you want me to log more of the RAFIG/airflow PIDs or more of the CALC.VET PIDs? Getting STFT and GM.VOLTS isn't an issue, I want to make sure I don't get more than 24 channels going at the same time.

[statesman]

+1 what joecar said about cleaning the MAF sensor.

The only airflow PIDs I need are MAF Frequency and MAF airflow.... so I'd say keep the CALC.VET PIDs and drop the RAFIG PIDS.

[pinstripebob]

I cleaned the MAF, did a RAFIG log and updated B4307 then went on a trip from Indy to the Tail of the Dragon with a group of friends. I got 3 logs on I-74, and like a dummy forgot to log GM.VOLTS. These logs wanted to increase B5001 and B0101, nothing crazy though. The logs are below:

Logs without GM.VOLTS, with cleaned MAF and updated B4307:
22490
22491
22492

When I got to our cabin, I updated the tune with this data, and of course my wideband decided to start giving wonky numbers. I have an analog Speedhut AFR gauge hooked up to my LC-2, and it was acting very erratically. Finally yesterday I recalibrated the wideband sensor and went for a drive. This time I got some good data with GM.VOLTS.

It looks like voltage dropped from 14.0-14.1 to 13.6-13.8 when the vehicle was stopped, you can see this at frame 23,847 and again at 24,394. This tune and log are below:

Updated tune and log with GM.VOLTS:
22493
22494

Data for B0101 and B5001 didn't look too bad. Volts stayed between 13.6 and 14.1V, which seems fine to me. To be transparent, I have never logged volts before, so I can't say much to its effects. This log started on the highway, then had a little in-town driving.

Overall there is the occasional bucking at very low TPS or off throttle, but it seems to be happening less often and when it does it is less severe.

I spent some time in heavy bumper to bumper traffic which SUCKED, but it gave me time to think about why my idle kept swinging wildly after I'd move a few feet using mostly only the clutch. I updated my spark underspeed and overspeed error tables to reduce their effects, and that got rid of almost all my idle swinging which is pretty cool. I don't think I mentioned that earlier, but it doesn't seem to be an issue anymore which is nice.

[statesman]

It sounds like you've got most of your issues under control now. Your voltage looks good. You might want to spend a bit of time working on your main spark and idle spark tables. There's probably some gains to be had there... just be mindful to keep spark variations small in your transitions between idle and non-idle conditions.