VandykeT/A

I understand that when tuning in lambda that 1 lambda = stoich for the fuel you are using and afr doesn't matter anymore. My question is since the wideband I am using (NGK/NTK) is programmed to read stoich as 14.7 wouldn't that cause it to display lambda 1 when I'm not actually at stoich for E10 or does what the wideband assumes is stoich even matter when I'm tuning in lambda and stoich is 1.00 no matter what. Basically I'm having trouble understanding if the wideband believes a fixed afr is stoich how will changing fuels still allow it to display 1.00 with E0 E10 etc.

JakeFusion

iTrader: (7)

Your wideband is tuned to believe lamda is 14.67:1. However, when you run it in HPTuners (for example) you set it for whatever the fuel is with the correct formula for the wideband input. Then lambda = whatever AFR is stoich for that fuel. As long as stoich is set in HPT and the input for the wideband the same, your ratios will be the same in all your tables.

The issue is the wideband gauge will still ready 12.1:1 as rich and 16.1:1 as lean even if you've programmed everything for E85. But HPT will display the correct AFR based on the correct lambda and ratio commanded.

For E10, if you're running 14.14 as stoich and lamda, then you can still read the AFR gauge. For example, straight gasoline would show 13.1 is a bit lean and 10:1 is pig rich. If you hit either of those, the real AFR is not that, but you'll know you're essentially rich or lean going off the E0 gas curve for stoichmetric ratios.

VandykeT/A

So I still need to adjust stoich in the tune to 14.14? The car was done professionally and its still at 14.67. I'm still not understanding completely. If I'm running E10 and I don't change the stoich in the tune how does tuning in lambda make up for that? Once the wideband is set up on a lambda scale will it automatically read stoich as 1.00 for whatever fuel I'm using? I understand the wideband box itself will still show a random afr because its set up for E0 but once its set up in lambda in hptuners will it automatically find stoich for the fuel and I just tune to 1.00 and ignore what the wideband box says? Basically does the wideband and hptuner (in the tune) need to be told what the stoich of the fuel is?

JakeFusion

iTrader: (7)

Yeah pretty much. You don't have to. Just if you get gasohol at 10% you're potentially off that much. Some just cut the difference and do 14.34 or so for E5.

HPTuners has the formula you need to run for E10 in the scanner for the wideband. If you want to run 14.14 I have the value for the NGK/NTK setup as I run it in the scanner.

Basically, you have to setup in the tune and in the scanner and make sure they are the same. It doesn't really matter what it is as long as it's the same. Then lambda = the same between the scanner and the editor.

VandykeT/A

So as long as the tune and the scanner are set to the same stoich ratio I can technically leave it at 14.67 and tune everything to 1.0 and it not be a big enough difference to matter? But actually setting the stoich close to the fuel being used makes it more precise correct?

VandykeT/A

Another question. How will the change in fuel effect the PE table. Before I started trying to tune I always thought 12.5-13.1 was a safe N/A afr.If I change the stoich does the same thing apply to top end. For instance if say I change my stoich to 14.14 and tune everything in the VE table perfectly and then command a 13.1 in the PE table will I actually be potentially half a point lean? I'm still confused. I understand in closed loop the narrow bands will shoot for stoich how do I know what to shoot for in the PE table without a dyno?Here is a copy of the tune in the car now. They commanded 12.6 across the board then a 11.2 up top if I change the stoich and fix the VE being a mess up top will I need to richen the 12.6? I think I'm over analyzing haha. jhr tune trans am.hpt

mrvedit

As you are figuring out, the wideband really only calculates lamba values regardless of the type of gas. The A/F ratio is just a "made up" multiplier that converts lambda into a number which many humans are more comfortable with.
Don't bother changing the "Stoic" value in the HPT, it doesn't change the tune any; it just changes the multiplier it will use when displaying A/F for you. IMO changing it just makes it more confusing.
So when someone says "My A/F ratio is 13 to 1", think of it as "If I were using E0, my A/F would be 13, but since my fuel really is Ex ('x' might be unknown), it really is something else and I don't really care what the actual number is because it doesn't matter". If that confuses you more, delete that sentence from your mind. This is why many tuners only talk about lambda.

Going back to your PE table, forget A/F ratio! In the PE is a number which will be divided into Lambda 1.00. So if your table value reads 1.16 that means your lambda is 1/1.16 = .86. If you were using E0, that would come out to 12.6, but since you really don't know the exact ethanol content of your gas, it is something else, but who cares because it doesn't matter. If you say your PE A/F is 12.6, we all know what you mean.

It is true that you cannot determine the perfect PE table without a dyno, but lots of people have tested these engines and know what the ballpark values are. It is typically 1.16 or 1.17 before torque peak and a little leaner like 1.14 or 1.15 after torque peak.
Without a dyno you can tune the PE table for the maximum trap speed. I have found it to be 1.16-1.17. If I go 1.15 or 1.18 I notice a drop in trap speed.

I hope this helps and doesn't confuse you more. In short the wideband records lambda and the PCM calculates everything according to lambda. A/F ratios like "14.6" are just a multiplier that HPT uses to convert lambda into a human-expected value.

VandykeT/A

I think I'm understanding. Basically lambda 1.0 is stoich and no matter what lambda 1.0 is lambda 1.0. I don't have to tell the wideband that lambda 1.0 is x afr. It reads the oxygen content and knows the stoich of the fuel that's being used right? If that's the case how do I know the lambda it shows in hptuners is correct when my wideband is displaying afr.

JakeFusion

iTrader: (7)

As long as stoich is the same in the tune and the scanner, yes lamda is 1.0.

If you're stoich is 14.67:1 then it'll be close to what the display shows. If I recall, it actually uses some other goofy calculation 14.55 or something for stoich. So they will be off about that much. Just pay no attention to the gauge of the wideband. Only what HPT says.

This is the beauty of tuning with lamda or EQ ratio - it doesn't matter the stoich. The EQ ratio (which is what the PE table is shown as) is always going to be about 1.16 at peak torque and 1.14 at peak HP. It's just what the engine and any type of fuel seem to like. Lamda is shown as something like .88 and EQ shows as 1.14 for example. EQ is divided by the stoich to give you the A/F. 14.67/1.14 = 12.86. Or with Lamda, you multiply by stoich 14.67*.88 = 12.90.

Get Greg Banish's book. He explains this pretty well. And tells you no matter the fuel, .90 seems to make best peak power, etc.

Btw, this is my PE ratio: 0 to 7200

1.0000 1.0000 1.0000 1.0000 1.1396 1.1396 1.1396 1.1396 1.1396 1.1396 1.1436 1.1611 1.1650 1.1611 1.1436 1.1396 1.1396 1.1396 1.1396

RonSSNova

iTrader: (3)

Your narrow bands will set the mixture to stoich. That's what they do.

So a good double check is to make sure your WB reads lambda of 1.0 when in closed loop during cruise conditions.

One of my AFX was .02 off. The other spot on. I simply changed the user defined formula by .02.

FWIW, I still like to convert to AFR in my head so I know I have my WOT fueling where I want it. Takes forever to get comfortable enough with Lambda numbers other than 1.0.

The really nice thing about using lambda when open loop under idle/cruise conditions is that at a glance in the chart display, you can read lean/rich directly in % error.
1.03 = 3% lean.
.97 = 3% rich.

Ron

VandykeT/A

I'm just not understanding how that if the stoich I have entered in both (I know they have to match) isn't the actual stoich of the fuel I'm using how will lambda on the PE side be accurate. Say for instance I set the stoich in the tune and in the scanner to 14.67 but the actual fuel I'm using has a stoich value of 14.14. My scanner will be showing lambda 1.0 at 14.67 and I understand the narrowbands will pick up the difference in closed loop but wouldn't that cause a .90 lambda (on the hptuners display not in the PE table itself) in PE mode to actually be lean.

JakeFusion

iTrader: (7)

That's assuming you have E10. Sometimes you have E3, E8, or E5. This is why you set your lambda to stoich of the fuel you use so you're PE table is on target - but it's hard with the ethanol blended fuels.

Some will say not to mess with it for E10 since it's less than a 5% difference. But I changed mine to be worst case. If I run E3 it'll be rich. E10 it'll be fine.

Others split the difference and put in something for E5 for stoich. Then you're barely rich or barely lean no matter the gasohol rating.

The stock formula in the scanner for the NGK wideband is V/.714 + 9. For E10 it is V/0.7388 + 8.742. Then you set to 14.14 in the tune. Then Lambda is set for E10.

VandykeT/A

So I can enter that formula in under user defined. Tune everything to 1.0 and worst case I'm slightly rich. It all makes sense now. Another question. How did you figure out the offsets for each. All I can find is the tuning school spreadsheet that enters .5-4.5v and 9.0-20.0 and it gave me something like .3636364 and 9. I appreciate all the help.

VandykeT/A

I have the NGK/NTK AFRM is that what the formula is for?

JustAFooL

iTrader: (45)

Thanks for this post and all the replies! I've just started tuning my car and I've had to read this thread multiple times but every time I read it it makes more sense. So.. What is the typical E value for 93octane?

JakeFusion

iTrader: (7)

Yup. It's what I have.

JakeFusion

iTrader: (7)

E value? E10 is the ethanol blend that's mandated. But it's usually some weird mix. 5% ethanol and 95% gas.

Straight gas is waaay better than the gasohol ****.

VandykeT/A

Ok I think I've got this understood completely in my mind now. When you say the stoich in the tune and the stoich in the scanner has to match your talking about when I plot commanded_lambda in the scanner. I was thinking the wideband still had to be set for a specific stoich. Let's say for instance I enter in your values for the offset and I'm using e10. stoich at 14.14 in the tune and when I start the car in closed loop I get 1.00. Then I drain that out change the stoich in the tune to 14.67 and use pure gas I should still see 1.00 on the scanner without touching the voltage offset or transfer function whatever it's called under user defined right?

JakeFusion

iTrader: (7)

No, that has to be changed back to the original values in the scanner. You can adjust the O2 sensor to pick up rich or lean conditions based off the stoich... but it's a voltage reading.

Remember, the stoich changes for fuel. It's like 9.7 for E85. The voltage is all HPT is reading from the O2. You have to adjust the voltage in the formula for the type of fuel being used.

VandykeT/A

I got it now. I couldn't grasp why lambda made it easier with different types of fuel if it didn't pick it up on it's own but now it makes perfect sense. With different fuels the desired afr will move around vs lambda where once it's set for the fuel the numbers are the same. Thank you. I'll use your numbers and change the stoich.