Stop bad fuel economy, eliminate idle and off-idle surging, regain that stock feel

[MNR-0]

Background:

I got a new cam installed recently. Since then, my fuel economy has been abysmal and Ive been plagued by reverse surging and slight idle stalling when coming to a dead stop hard. It runs rich on startup and consumes 4.7L/HR at idle and in gear, 4.5L/HR in PARK.

Last tank lasted 240kms, or 140 miles.



Diagnosis:

I concluded the problem must be proportional fuelling. The cam is closest to an F13 and I run LTs. I knew the O2s would cause issues being farther away from the manifold but not this much. Even cruising, they looked sick.

You can't trust O2s with large cams, especially with any significant overlap beyond stock. I have 8* positive overlap at 0.050.



Solution:


Part 1: Engine re-cal for open loop and O2s at idle and off-idle


Disable closed loop at idle and part throttle by setting my PE TPS vs RPM to 0 from 0RPM to 1200RPM inclusive

Set PE enrichment to 1.0 from 0RPM to 1200RPM inclusive

Raise all O2 rich/lean vs. airflow switchpoints by 40%. I set mine all to 541, as I wont be using Airflow modes 0-6 since I will be in open loop.

Set Closed loop proportional O2 error to:
0.06250 0.06250 0.06250 0.06250 0.06250 0.06250 0.06250 0.15479 0.18750 0.21484 0.24023 0.27002 0.27979

My stock settings were all 0.175, which I reckon caused too much gain. The above settings are stock for Closed loop proportional idle O2 error, so I just set them to be the same.

Save and Flash into the car.


Part 2: Dialling in idle AFR


Now for some live tuning. You don't need to use a wideband either.

Connect the scanner and startup the car. Turn on the scanner, opening up the VCM bi-directional controls. Remember that right now you are running in open loop.

Disable Adaptive spark control

Check your Instantaneous fuel usage in PARK. Mine was 4.5L/HR

Check and set the AFR to 14.7. Raise this by 0.5AFR and check your instantaneouds fuel usage. Also listen to idle quality. Keep raising this by 0.5AFR until the car begins to chop and lope. This means you have reached a lean condition and should stop there. Time to back it off. Back the AFR off by lowering it 0.1AFR at a time until you have found the sweet spot for idle quality. Check your instantaneous fuel usage - it should have significantly decreased.

My sweet spot was 16.3 from 14.7. My instantaneous fuel consumption dropped from 4.5L/HR to 3.6L/HR.

Turn off the car. Go to your VE MAP and multiply your left 3 columns (all of them - 400RPM to 1200RPM inclusive) by your sweet spot equivalence ratio. For example, 14.7/16.3=0.9. Blend this into your next 3 columns.

Save and flash.

Startup the car and repeat.



It took me 2 goes - all of 10 mins to completely get rid of all my idle and off-idle issues. No more reverse surging, no more fuel stink on cold start, smooth return to idle characteristics.

Remember, your O2s are useless with large overlap cams so ignore what your LTFT/STFTs are reading. A large cam will ALWAYS run lean at idle and off-idle - according to the O2s.

Infact, I turned off my LTFTs all together - permanently - 1 year ago.

My thinking is the larger the cam the more you need to rely on open loop and stuff the O2s. Also note that your VE (MAFless) tune should be dialled in to begin with. I could run open loop all together but since I dont have a wideband I dont want to risk it. With my size of cam, it begins breathing efficiently approaching 2000RPM anyway so I chose to only go to 1200RPM for open loop.

Note that this is effectively a hybrid open/closed loop approach. You can have both - after all.

My fuel economy after a 100km round trip dropped from 24L/100KMS to 16L/100KMS. The car is much smoother off and return to idle.

All feedback welcome.



Special thanks to Martin Donnan. Thanks mate.

[MeentSS02]

This sounds all fine and well, but I still don't understand what the airflow modes do. I run full time open loop, so most of this wouldn't apply to me, but I'm still curious.

[chrisj6]

i am wondering about this also, i noticed the '01 and '03 Silverado 5.3L closed loop vs. airflow modes are a bit different, and I was wondering if i could gain some fuel mileage if I changed this.

Chris

[jimmyblue]

Airflow mode vs airflow just makes you some "airflow
cells" for closed loop to index stuff simply. Like the
fuel trim cells, sorta. Being able to control the detail
of the fuel dithering, accommodate lazy or depressed
O2 sensor output, etc. is nice. Keeping the system
from swinging so far over center has to improve the
economy (like, you average out to proper fueling but
spend very little time there in fact; statistics, ptui!).

I'm going to have to apply this to my car eventually.

[MNR-0]

To add to Jimmys post, if you are in open loop you dont use closed loop fuelling at all so airflow modes, proportional fuelling etc. do not apply.

Ariflow modes simply index ranges of dynamic airflow. For example, DA 10g/sec indexes to 4, DA 14 indexes to 5 etc..

This is all determined by the setup in the Base Airflow Mode table. Here is an example:

00 04 08 12 16 20 24 28 32... dynamic airflow (g/cyl)
00 01 03 05 07 08 09 10 11... airflow mode

The airflow mode allows the PCM to cross reference all its closed loop fuelling parameters to dynamic airflow. So at idle, I might be pulling 10g/cyl. This cross-references to airflow mode 4, which in turn corresponds to either a O2 error gain value, O2 proportional base value, O2 switchpoint value etc...

[SSUte01]

I am following your advice to run the OL idle following the installation of the cam. I am however having trouble in two areas

1. I do not seem to be able to command any other than 14.63. I have the EQ ratio at idle MAP and ECT to 0.95 (AFR ~15.5) to see if this helps. However when I log AFR commanded it remains at 14.63 following startup adder. I thought it must be because I needed to se the PE ratio to 0.95 at idle rpm but still goes to 14.63. I then did a gross error check set the EQ ratio to 1.5 and momentarily the commanded AFR went to ~10 and all of a sudden went back to 14.63

What have I missed - I am very perplexed.

2. The other area, I believe this is simply due to my rich idle state when the fans kick in, it will hunt up and down for quite a while prior to settling (sometimes it wont settle at all prior to fans switching off). I have tried several additional g/sec load from v.small 0.17g/sec right through to max.

As I said I think this is a by product of the rich idle, but I could be wrong.

Apart from this I am very satisfied with the tune, timing, PE, VE, DFCO, I am just stumped at them moment.

Cheers

[EC_Tune]

The PCM will always pick the richest of the air fuel/EQ ratios. So if the "Richest" value you have is Stoch at 14.6x you will get 14.6x in PE Open Loop.

EC

[SSUte01]

I thought that about the OL AFR so I went ahead and set the PE table to 0.95 at 800rpm as well as leaving the EQ ratio table set to 0.95 at idle MAP and it still remained at 14.63 following the startup adders.

Oh well, not to worry used the live controls to make some slight adjustments to the timing and the VE Mult and set the additional air for the fans again and found the sweet RPM and not she's all apples. Can hardly notice the fans come on - I think that there is still some very minor fiddling to do with the idle VE but it is pretty smooth.

Thanks alot for your help

[MNR-0]

One thing you need to watch out for is whether you are infact in open loop. If your PE enable MAP values in the PE WOT section are higher than your idle MAP you wont see open loop at all. If you are seeing 1.0 equ. ratio in the scanner and you have commanded 0.95, then you are reading the FA ratio from somewhere else - likely you are still in closed loop. Go through everything to make sure you are not in closed loop.

[JimMueller]

What is the custom PID for instantaneous fuel usage? The PID I have is for instant MPG, but since I'm idling it always shows 0.

Edit: I removed the [sens.20] from the formula and now it's non-zero and follows TPS. It's hovering around 1.7, and bumping the AFR from 14.7 to 16.7 only swings it about 2/10's. That's why I'm concerned my current formula is wrong. I was hoping for something which showed a larger swing... plus, even at 16.7 it's not really changing the lope. Pulling 5 degrees of spark makes it really lope, though.

[MNR-0]

I just used the dash fuel usage readout.

[SMOKINV8]

Nice write-up, I'm looking forward to trying this out. I've got a question about what I've quoted below. Why do you go in and edit your VE table, rather than modifying (reducing) the appropriate cells in the OLFA table? I'm pretty sure it would give you the same result, but I want to make sure there's not a method to your madness that I've overlooked.

Turn off the car. Go to your VE MAP and multiply your left 3 columns (all of them - 400RPM to 1200RPM inclusive) by your sweet spot equivalence ratio. For example, 14.7/16.3=0.9. Blend this into your next 3 columns.

Save and flash.

Startup the car and repeat.

[MNR-0]

Because the VE table is the base measure of airflow for all OLFA parameters. Catch them all at the one place. If you are running a MAF, then you need to do the same thing to the low airflow values too.

[silverTA2002]

Ok, so just to make sure I understand.

After all the logging is said and done, the only PERMANENT changes that you make to your tune are to the VE table? (i.e. set all the PE stuff etc. back to what it was before this procedure?).

If so, couldn't you just ignore the LTFTs at idle and lean out your idle a little in the VE table until you don't have a rich idle problem? (And then lean out MAF calibration at idle if you're keeping it).

[silverTA2002]

Also, if it's just VE changes, wouldn't you only want to change the non-load/idle values in the VE table, rather than the entire 1200 rpm column (for example)? 16.3:1 might be fine for idle and returning to idle. But, that's not necessarily where you want to be for 60mph in 6th gear.

[SSUte01]

Sorry to go off topic a little but I use the custom PID but take 236 and divide whatever the PID is to get L/100km (Aussie)

Question about the instaneous fuel (if you dont have a readout). If you were to multiply IFR by 3600 to get g/hour and then multiplied this by the density of the fuel approx 750g/L (you could use the section of the MPG PID to get this more accurate). Your result would be max possible L/hr. Now my question is - if you multiply this figure by Inj DC will this be an instantaneous fuel readout in L/hr or am I way off. I did an idiot check in my head based of my figures and I got approx 3L/hr.

Can anyone clarify this for or am I

[MNR-0]

Also, if it's just VE changes, wouldn't you only want to change the non-load/idle values in the VE table, rather than the entire 1200 rpm column (for example)? 16.3:1 might be fine for idle and returning to idle. But, that's not necessarily where you want to be for 60mph in 6th gear.

I think you are getting confused. Any change you make to any calibration map is permanent until you change it again.

Do you cruise at 60mph at 1200RPM? I have a stalled auto. There is no way I see 1200RPM apart from flashing through it when moving from a stand still. I cruise at 60mph at 2200RPM.

Its all about off-idle and return-to-idle tuning here - not cruise - and I dont want this thread to go off topic where this can be answered in other posts.

Just follow the procedure. You can, if you want, just alter the VE calls at idle by logging which ones they are. But I think you will find it fluctuates quite substantially with a cam. In the end, the only cells you are worried about are the idle cells. You wont see 90kPa at idle unless you have a sewer pipe of a cam - so it not an issue to modify the entire column. Alos, you dont want peaks and troughs in your VE - smooth transitions.

[silverTA2002]

I think you are getting confused. Any change you make to any calibration map is permanent until you change it again.

Do you cruise at 60mph at 1200RPM? I have a stalled auto. There is no way I see 1200RPM apart from flashing through it when moving from a stand still. I cruise at 60mph at 2200RPM.

Its all about off-idle and return-to-idle tuning here - not cruise - and I dont want this thread to go off topic where this can be answered in other posts.

Just follow the procedure. You can, if you want, just alter the VE calls at idle by logging which ones they are. But I think you will find it fluctuates quite substantially with a cam. In the end, the only cells you are worried about are the idle cells. You wont see 90kPa at idle unless you have a sewer pipe of a cam - so it not an issue to modify the entire column. Alos, you dont want peaks and troughs in your VE - smooth transitions.

Ahh... this is where the difference between a manual, and a stalled auto car come in. I just did a 20 minute part throttle LTFT log. At 800 rpm, I hit cells from 30 to 90kpa. At 1200 rpm, I hit cells from 20 to 100kpa.

So, if you're an M6 car, you shouldn't modify the entire VE column, but just the idle/return to idle cells. Then hand smooth to make sure you don't have dips in your table (same rules as regular VE tuning).

[silverTA2002]

I think you are getting confused. Any change you make to any calibration map is permanent until you change it again.

Right. But in your instructions, you don't mention putting VE and OLFA back to stock after you're done logging. My "permanent" question is asking if you should put those back (instead of 1.0). And if you put the o2 switchpoints back to stock, etc.

[possum22]

And if you put the o2 switchpoints back to stock

I think the PCM ignores them when you are in open loop, I had mine 361 to 375 in the first 3 columns.