Show you what where? You open the math formula. You add a term. Do you know what a term is? I have been a math tutor for 12 years if you need help I will teach you these terminology. There is no such thing as subtraction or division btw.Originally Posted by SiriusC1024
Remember this? This is how you do it. You make a math for each offset region. My car has 3 offset regions so I have 3 math formulas.
If you give me a log with narrowband and analog wideband I will point out the relationship between regions and show you how to adjust offset for each region and bake math formulas for that specific example into a math equation(s) with the caveat understanding that every log and every vehicle will have its own math and its own offsets, even the same vehicle from day to day will vary. Everytime I make a log I always adjust my offsets that very same day as needed to make gauge agree with log otherwise its useless fucking waste of time.
all around the world people know
https://www.miataturbo.net/megasquir...e2/#post826213
https://forums.nasioc.com/forums/sho...4&postcount=14
The wideband is correct. The gauge will agree with the narrowband. Either cheap wideband or factory narrowband could be slightly off but which one is off? You will never know with just those tools. Many factory high power cars with turbos have widebands already, narrowband is obsolete where applicable. There is a difference in sensors and like anything how you use it is important. And getting a clue in your mind of the true fuel flow for the power and fuel quality and application is far more important than any wideband reading anyways.
https://ecotrons.com/accurate_lambda...su_42_sensors/
The newest best aftermarket ECU all use exclusively widebands with digital communication like the new OEM vehicles that also have widebands. The only reason GM stuff doesn't have a wideband from the factory is COST. MONEY. They are superior in every way to narrowband and will absolutely perfectly agree with narrowband reading if used properly. The wideband is a narrowband inside it, same thing just more advanced. L@@K at the logs the guy is posting here from his CANBUS wideband.
https://forum.hptuners.com/showthrea...reading-lambda
Every single log 99 logs the narrowband matches the wideband. There is no accuracy issue with the wideband. Only a tuner issue when they don't know how to adjust analog offset in their logs via grounding and math formulas.
Last edited by kingtal0n; 12-23-2023 at 10:48 PM.
- If you need more than 1 offset that means it's not an offset problem. It's a calibration one.
- If you're moving the wideband toward the narrowband fuel trims that means wideband is inaccurate.
So why not just tune CL in the first place. Why all this tune with wideband nonsense you won't stop talking about? You're just adding a layer of inaccuracy even if the wideband is worked toward what narrowband CL trims indicate.
As much math as you say you know you're not very logical. I'll leave you this to figure out. Offset is applied incorrectly in that formula you talk about. It's a voltage offset, right? See the problem?
Vegr/((Vmax-Vmin)/(AFRmax-AFRmin))+AFRmin+AFRoff
I pulled up a random log from the gentlemen with canbus AEM wideband and 1 working narrowband. The wideband feeds on both banks but the narrowband just from 1. The injectors are dirty as hell he shows a picture they are gunked up like crazy. But despite these things the narrowband and wideband correlate extremely well.
low rpm
mids
a bit higher rpm
Notice No analog offset. No accuracy issues. The two agree competently pretty much all the time.
Now that I am paying extremely close attention I do notice the narrowband has some indecisiveness intuitive to a sensor which produced its own voltage and relies on exhaust gas sampling. The swirling, changing mixture makes the narrowband a bit wishy washy for a couple hundreds milliseconds sometimes on some flip flops. The wideband on the other hand is programmed well enough by AEM (it matters) that it does not show indecision and lines up with the injector pulse expected change, and it never fails for the narrowband to catch up after a split second. Basically the narrowband is a childs toy compared to the wideband, as I knew all along but now even more certainly. A problem I forsee for most people they can't see the kind of patterns in a signal or device that I can and maybe interpret sometimes what is really a digital signal bounce as some kind of error or inaccuracy, you must never forget this sensor A. producing its own voltage B. relies on electrons flowing into a base and C. has no de-bounce output circuit I am aware of, the true signal can be seen in the wideband while the narrowband plays with itself on those occasions.
Offset changes with RPM and LOAD so you always need multiple offsets. Calibration is never to be done especially on AEM widebands and especially if they offer you a calibration feature.
No it means the ground is different for the analog offset. A canbus wideband will never disagree with narrowband the wideband is perfection incarnate. The issue is you fail to account for grounding when using cave man analog input.
- If you're moving the wideband toward the narrowband fuel trims that means wideband is inaccurate.
1. Wideband tells me everything a narrowband could tell me and more. Why do I want extra holes in the exhaust that is just ridiculous. I custom weld all my exhaust before you say it already has holes.So why not just tune CL in the first place. Why all this tune with wideband nonsense you won't stop talking about?
2. Wideband allows me to target lean air fuel ratios 16:1 which maintain clean cylinders and reduce carbon deposits on plugs and improve economy and reduce oil washing down carbon chains (improve oil quality).
3. Theres more but who cares. This is how all aftermarket ECU do things. There is no narrowband. I want my $25 piece of shit OEM Gen3 computer to work like a $2000 aftermarket computer this is how its done.
Actually the narrowband is the layer of innacuracy because it makes its own voltage and is subject to the natural offset between the ECU ground and the ground wire that feeds the ECU from the narrowband- which not all vehicles have! There are such things as 1-wire narrowbands, after all, which rely on grounding through the exhaust. If you clean the exhaust on those the narrowband will read differently! Holy sheet bro. I've been using narrowbands much longer than you and wrote applications around them, I think i know what they can and cannot do.You're just adding a layer of inaccuracy even if the wideband is worked toward what narrowband CL trims indicate.
As much math as you say you know you're not very logical. I'll leave you this to figure out. Offset is applied incorrectly in that formula you talk about. It's a voltage offset, right? See the problem?
Vegr/((Vmax-Vmin)/(AFRmax-AFRmin))+AFRmin+AFRoff
First of all there is an offset in the Vmax and Vmin. We or at least I do not know the minimum or maximum voltages that one may input to the EGR I always assumed it was 0.5v to 4.5v but recently because of all this nonsense I measured it could actually report 0v and a high of around 4.28v around that from my wideband. In any case there is some embedded offset in every voltage number you write as a math term, there is no integer. So anytime you see an integer, forget it. Its a fudgator.
Second of all The sign + means you can add them. So for example if AFRmin is "10" and AFRoff is "0.1" you just add them together to make 10.1 or 9.8 or whatever. It the number of volts the offset is off between the gauge and the ECU. You can measure with a multi meter by taking + to the analog output and - to the gauge ground and write down 1 value. Then take + to the ECU analog input and - to the ECU ground and write down the new voltage value. The difference is going to show up in the log file as the analog offset between output and input to the ECU minus whatever the input diodes consume and all that crap inside the computer comparators and shit.
Better stop me I'm welding up pipes without a spot for a narrowband...
I'm also welding little pipes onto my maf sensors because I want to fit larger air filters
And these cars are MAF ONLY so figure that one out.
oops I converted this one to MAP only because the customer wanted more power and nobody wants a piece of shit in the airflow path when trying to actually make power.
Holy Autism. WTF is voltage offset if the WB is grounded with the same ground as the ECM is "feeling?" Idk what your hatred for NB's are. Maybe you can explain voltage offset vs guage, vs ecm, vs the input to the MPVI? I am soooo confused lol. If you have to skew the voltage vs lambda, to get the reading you want... then the WB is inaccurate?
I already explained it here many times. But one more time is fine. The wideband is perfectly accurate. The gauge is reading perfectly. There is no issue. You can tune the car on the gauge fine. Did it for 20 years without logging.
It will always be right if you know what you are doing, and understand the limitations of a cheap wideband in terms of power-pressure-temperature situations. There is no analog offset if you want to read the wideband gauge and tune any car. This the wideband is accurate and calibrated etc...
The problem comes when you try to log the wideband without a digital communication. If I set something up to output 5v and it has it's own power supply then the 5v it provides will NEVER EVER match the 5v of some other ECU or microcontroller. Its impossible for them to be on the same level consistently. Even if you tie the grounds together there is a difference in the current flowing from each device which, on an automotive application, varies with load, temperature, even time of day due to environmental variables which influence sensors, battery, charging, even the ECU temperature and the temps of the wires. Therefore, you can either use a canbus communication like all aftermarket 4000hp capable ECU use, OR, you can fuck with the analog offset to make the two match as this 2007 thread describes.
https://forum.hptuners.com/showthrea...oltages-How-to
If you missed it, that thread from 2007 shows you how to setup the scanner so that the GAUGE (which is always accurate as proof the wideband is working accurate and calibrated) will match the scanner (which is NEVER ACCURATE when using analog offset).
In order to properly tune a vehicle using a wideband you must L@@K at the GAUGE when tuning and go by what the gauge says, not the scanner. When you review the scanner later say, at home in bed, you must remember or write down the difference between the gauge and the scanner at every load/rpm increment, at least remember how far off it was, otherwise its pointless and useless to use those wideband logged values.
I don't hate narrowbands I just don't see the point and neither does any aftermarket ECU company in the world and many OEM manufacturers have eliminated the narrowbands up front as well. They are still used as a cheap way to ensure catalyst function on some vehicles I am sure of that, but not for tuning or optimization and especially not for tuning a high performance vehicle these days. Seriously guys I know this forum is in the dark ages but get a fucking clue. Even your own forum's catalogue ECU is going away from narrowbands and moving to widebands soon. Wake up!!!
I've logged wideband values on a computer called Power FC for Nissan Skyline and Toyota Supra,
It uses the same analog input and has the same issue with an offset voltage that must be calibrated by the end user.
Perhaps this is why I have such an advantage going into this discussion because I've used this ECU and dealt with the offset issue for many years already.
Suffice to say it doesn't matter how you intent to capture the voltage of a wire on any analog pin into any microcontroller, there will always be some offset unless you get really creative with the design.
Its easy for me to remember "Ok around 3000rpm the wideband was about 0.2 richer on the gauge than the scanner shows today" so I don't have to stop and change the math while trying to drive 80mph.
But if I don't even have my laptop I could still recall the air fuel ratio display on the gauge at some specific pressure and RPM anyways, and even write them down on a scrap of paper. I don't get what you guys are doing with all these maths and formulas and sensors and crap you really overcomplicate tuning that way. You only need 3 points on a VE table to make the VE table nice and smooth and damn close. I don't envy any one of you but I wish you would just think and listen for a change so you can make it easy on yourselves.
Why do OEM still use widebands up front in their brand new cars , they are more expensive
For so many years they use widebands from the factory in these cars
Since... how old???
I don't know why they are using widebands in all those old and very new expensive cars. The wideband sensors cost more. Maybe there is a reason... huh? Whats that? All modern high performance ECU also use widebands? Say it aint so. Shit. So both factory and aftermarket high quality ECU are using widebands since the 90s? No way!!! Whats that on the floor... some clue? Getting some clue!??? hahaha
There is a reason I don't use math for tuning. It is because tuning is not math, it is art.
Yes, and as I explained the wideband in OEM applications is part of closed loop. It's not OL wideband the way you do it. Closed loop is the key here. That's why I say narrowband fuel trims.
Tuning is art. Ah. Last I checked the tables in a tune were numerical, bud. You know what an array is?
You contradict yourself so much just to get out of admitting you're wrong. Literally two hours ago you're talking about offsets to math formulas. Now you're like, "That doesn't matter my artist's intuition exceeds this reality."
https://en.wikipedia.org/wiki/Time_Cube
Cognitive dissonance. Do better.
Exactly.
How can you possibly think the wideband is inaccurate when the gauge is reading the correct value and agrees with the narrowband all the time?
I will bet you my account vs your account loser deletes it, that if I install a narrowband behind my wideband they will both agree all the time on my gauge.
numbers are an illusion. Mathematics is a convenient way of lying.
Wideband doesn't agree with narrowband fuel trims in all situations. I have an example of this happening Post #2 here:
https://forum.hptuners.com/showthrea...nd-vs-Wideband
You can't explain it other than saying it's an offset problem. You can't explain the math to correct the offset problem. Then suddenly you say math doesn't matter. The very formula you recommended doesn't matter because now you're a fucking artist. Your story keeps changing. It's kind of ridiculous.
Yes. Hard to understand because you make it up as you go.
It's correct to tune like the OEM does it. In Gen 3 that means CL narrowbands. Others it's CL widebands. Not OL. Not OL, kingtal0n. Not. OL. Got it? Your method of ignoring the factory controls misleads people, and it's frankly retarded. There's always some excuse. Might work on newbies, but it doesn't work with people who know about this stuff.
Tune your car in whatever way you want. I don't care. Have the decency to not recommend it to others.
Last edited by SiriusC1024; 12-24-2023 at 02:32 AM.
Thats not a canbus wideband. There is no way to log a non-canbus wideband accurately. Its impossible. A non canbus wideband is only accurate while you watch the gauge which is what this whole thing has been about.
You can only take a video of the gauge in the car of the wideband which is absolutely perfectly accurate at all times and will always agree because the wideband itself is accurate. The signal going into the ECU is never accurate because it is analog.
Nobody can be this stupid so I think you are just trolling at this point.
What else could it be? You won't look at my canbus wideband which is perfectly matching the narrowband so you refuse to admit its your fault the log is wrong. Its YOUR fault you didn't math out the offset properly.You can't explain it other than saying it's an offset problem.
Perfect matching up between narrow and wideband oxygen sensors thanks to canbus communication using HPtuners. Perfect.
I explained it several times. So does the thread from 2007You can't explain the math to correct the offset problem.
https://forum.hptuners.com/showthrea...oltages-How-to
If you can't perform simple addition then... ? Failure to do 3rd grade math. I just don't get it. If you provide a log file I will create a formula for you. Even a non linear polynomial if you want to automatically scale for rpm and load. I can do any math. It is one of my strengths. That is how I know that math is an illusion.
I'm always an artist. I am always saying math doesn't matter. That never changed. I am simply reminding you because I am getting bored of your constant ignoring how a microcontroller works.Then suddenly you say math doesn't matter. The very formula you recommended doesn't matter because now you're a fucking artist. Your story keeps changing. It's kind of ridiculous.
I didn't make up that thread from 2007 and it has everything you need to be successful. https://forum.hptuners.com/showthrea...oltages-How-toYes. Hard to understand because you make it up as you go.
I recommend ignore everything I said and focus on that thread and you'll be fine and so will everybody else. I have no problem with you teaching people to log their wideband but you MUST INCLUDE That thread or else it is worthless and meaningless and you are teaching them wrong things. That is all.
The OEM tuning is done for cat converters optimization. If you removed the cats then it is no longer important nor optimal to maintain narrowband air fuel ratios. Instead, high performance engines target 15.2 to 16:1 for optimal engine performance during idle and cruise conditions. The only way to achieve this performance of a $2000 aftermarket computer is to use $25 OEM GEN3 ecu with open loop. We use open loop because without it the $25 ecu is just a piece of shit $25 ecu with no benefits for your high performance project since you are stuck using garbage air fuel ratio and tuning strategy.It's correct to tune like the OEM does it. In Gen 3 that means CL narrowbands. Others it's CL widebands.
I will teach the way I teach since I am a teacher and have the necessary credentials such as a doctorate. You on the other hand are not in any position to teach anything not even how to wipe your own ass. This thread is a great example of why you cannot be teaching anybody properly how to use their own equipment since you cannot even grasp the most simple devices such as a 0-1v sensor signal as a digital construct. If you were savvy you would understand that a wideband can function as a narrowband and perform closed loop operation for gen3 ECU without the narrowbands- beholdNot OL. Not OL, kingtal0n. Not. OL. Got it? Your method of ignoring the factory controls misleads people, and it's frankly retarded. There's always some excuse. Might work on newbies, but it doesn't work with people who know about this stuff.
Tune your car in whatever way you want. I don't care. Have the decency to not recommend it to others.
https://thedeltaecho.wordpress.com/2...p-afr-control/
But you don't realize these projects exist, do you? Your credibility evaporated the minute you provided somebody a formula for an analog signal without teaching them about the offset that incurs from all analog device inputs. Good day.It could also be used to adjust the AFR to 15.5:1 at pressures below 40 KPa and create a leaner idle.
Last edited by kingtal0n; 12-24-2023 at 02:56 AM.
Either way, only way you can see if the wideband is matching the NB would be under a steady state exhaust flow, and only at Stoich. I don't get your screenshot saying it aligns together but it is at a 13 afr and the NB is railed at 900?
We need to stop these sidetrack rants of misinformation for the newer members that are coming on to this forum to learn.
I just seen a thread about open loop vs closed loop. Like WTF, that shouldnt even be a question (unless different or extreme circumstances, which that was not) lol. I just figured this kind of stuff would of ran its course by now.
It hasn't. So many times someone new asks a question, kingtal0n takes advantage of their ignorance in order to preach, and there we go prepare for the irrelevant text walls. Good luck getting anything done at that point. Tired of it. I made that thread so that any time in the future all I have to do is link it as a reply to him.
Found some LSA injectors locally and scooped them up, definitely feels more responsive on the pedal with less transient feeling issues. All of my STFTs shifted lean and boost areas started shifting rich. Would the shift at low maps be due to it now running at the very bottom end of the MPW?
2004 Silverado 1500 Single Cab 6.5' Bed
6.0 LQ4 Stock Crank/Rods/Pistons
823 Heads - Stock
Cam Specs: 226/242 .600/.600 118+4
LSA Supercharger with CTS-V Lid, 2.45" Pulley
Walbro 450 Fuel Pump with Return Style Fuel System, Boost Referenced Regulator, LSA Injectors
4L80e with AWD Swap
1-7/8" Long Tube Headers, Y-pipe, Single 3" Exhaust
Reply With Quote
