Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[FLTRI]

Since this topic ran through 261 posts without 1 "tuner" determining if O2 sensor voltage @ WOT can be used to tune to a target AFR I have come here to see if it will spur valuable input by tuners.

Question:
If a tuner were to put a bike on the dyno and measure WOT AFR with his broadbands and record the voltage from the sensors...can that voltage number be used in reverse to get to the previously measured AFR?

If this is possible to use to get close to a dynoed WOT AFR for those DIYers who do not have the equipment it may provide a way for them to tune WOT after v-tuning to the 83kpa limit.

Hopefully this will prompt some good input rather than the BS crap that got tossed around even by the moderators on the "other" site.

Thanks,
Bob

[mayor]

If a dyno tuner where to help come up with a way to reduce or eliminate the wide open narrowband tuning restrictions for the DIY'er, wouldn't that pave the way to reduce some of the perceive need for dyno tuners?    I'm not real up on this tuning stuff, but I thought it was just fuel and timing, so if you take one of those away doesn't that limit what is actually needed when seeking a professional?  I think this is a great idea,  hopefully you will get some pro tuner participation here.   

btw what's this "other" site you speak of, got a link. 

[Fired00d]

If a dyno tuner where to help come up with a way to reduce or eliminate the wide open narrowband tuning restrictions for the DIY'er, wouldn't that pave the way to reduce some of the perceive need for dyno tuners?    I'm not real up on this tuning stuff, but I thought it was just fuel and timing, so if you take one of those away doesn't that limit what is actually needed when seeking a professional?  I think this is a great idea,  hopefully you will get some pro tuner participation here.   

btw what's this "other" site you speak of, got a link. 

I'm sure you are aware of the other site since you have an affiliation with it.. and it would be appreciated if you keep what goes on at "that" site, on "that" site. Thanks.

 
Ride Safe,
Fired00d
 

[Buckeye_Tuning]

HaHaHa.  Excellent!

I do NOT think it would take away any need from pro tuning.  The same guys that go to pro tuning would still go... there's a LOT more to it than getting VEs dialed in.  But...  I DO think this could help the members that will DIY their bikes, regardless.  The idea has been floated AND worked upon to make the NB sensors in our bikes used for 80+ kpa and would eliminate the need for 'extending' VE tables.

Mike, no need for pretense.  FiredOOd is cool Mod.  Just need to watch what he says about that other crappy site.

I have found, on my very own, that when using calibrations that are AFR cals... extending the VEs into the 90-100 kpa ranges works well.  Actually, It looked to me, during testing, the VEs came out around 5% too rich.  This using a TTS AND a Twin Scan.  But...  I have also found the same to NOT be true when working with Lambda calibrations.  Sometimes I find those to be lean.  Not good.  I would like to see DIY guys have the ability to check out this stuff and not rely on software alone.  From my own understanding of things, this very well MAY be accomplished using the NB sensors.

I can NOT talk for Bob at all.  But what I can do is lay out why I, me, would like to see a couple guys with Dynos do some tests.  I surely will.  The very FIRST idea we all would have to get past is the idea that the NB sensors can not read anything as rich as 13.5 AFR.  Once some go out of their way and show that it IS (or is not) doable, and repeatable, the next thing would be to get into the software itself to use this.

A LOT of folks have stated that the NB can NOT reliably measure 13.5 AFRs.  I disagree with those guys.  Some have made good arguments on why they will NOT work, but some of those ideas have holes in them.  For example, ... "The OEMs do not use them for fuel control, so we can't either."  Know what?  I agree with this statement 100%!!!!!  I do not think we could.  But.......

I am NOT proposing, and neither is Bob that we use these for fuel control at 13.5 AFR.  Nope!  Instead, all we would need is for a Steady state tuning run to show a single 13.5 AFR=13.5 AFR connection to things.  These sensors may not be accurate in 13.5 as fuel control devices that warble fuel rich, lean, rich, lean 8-50 times a second...  but what is to say those self same sensors wont read a bike that is held steady at a certain Map and RPM to gather data with?  I think NOTHING says that.

Both the NB sensors we all use and the Wide Band sensors we all use are Bosch sensors.  They actually work off of the exact same principles... a Nernst Cell, that measure the oxygen difference between fresh air and the air within the pipe.  Same ultimate sensor.  The real difference is the wide bands use an electromechanical 'pump'... this is nano tech stuff, to help insure fresh air to the cell and that makes WBs react faster outside of the 0-1 volts area the NBs work in.  WBs are cut to work 0-5 volts because they CAN, using this 'pumping' technology.  This allows an AFR swing of like 5-18 afr.

But... think on this a bit... if data was collected at a STEADY pace, giving TIME for the sensor to quiet down?  A nernst cell is a nernst cell.

Next argument is that a NB sensor ONLY reads 0-1 volt.  This is true.  And a WB reads 0-5 volts, this is also true.  Equally true is the fact that a WB is only used reading 1/100ths of a volt.  All of the WB controllers I have seen read in 1/100ths of a volt.  So.... a WB sensor uses 1300 'steps' to read voltages.  So, there are 1300 'steps' between 5 afr and 18 afr, with me?

A narrow Band uses millivolts... that is 1/1000ths of a volt.  So that sensor uses 1000 steps to read its voltages., these 'steps' are viable from .9 Lambda to 1.1 Lambda  ( 13.2 AFR to 16.1 AFR) [Bosch Documentation] Way finer degree of accuracy in a NB sensor than a WB sensor.  1000 'steps in THREE AFRs, instead of 1300 steps in 13 AFRs.  See the way ass finer degree of resolution?

And... the amazing thing is this...  our bikes are equipped to read a millivolt  of change in a NB O2 sensor!  Anyone that has DIY or Pro tuned has seen O2 sensor traces from the data collected...  it is in MILLIVOLTS!!!

I do NOT see a problem with reliably obtaining data at 13.5 AFR... and if one could find 13.5 AFR and input that into the tuning process, it's a no brainer that once that AFR has been met and recorded, one could alter that AFR differently without issue, to say 13.2... 12.8... etc.

So, hopefully some here on CVO will help determine whether a NB sensor can reliably read 13.5 AFR.

I propose doing this in an easy fashion on a bike, easy on the bike, also.  Set up a bike that outputs 13.5 AFR at say 3000 rpms and 60 map.  Use Wide Bands to do this.  This will not beat on a bike like WOT would, it's just a test after all.  Once the bike outputs like you/we wish, look at datamaster and see what millivolt reading is showing.  Simple!  It does NOT matter if we do this test at WOT or just above idle.  The test is all about the accuracy of the sensors.

If one THEN alters the VEs by like 25% richer and sniffs it, see what the AFRs are in THAT one spot.  Use the NB sensors and look at the voltages now.  I am sure one would see well over a volt.  like 1239 millivolts.  With me?

Now... manually lower the VEs in that test spot until you get the NBs reading under 1000 again.  Then manually lower ever so slightly until you MATCH the millivolt reading that the bike had at 13.5, BEFORE you messed with the VEs.  Once that is done, take the wide bands and see how close you are to the ORIGINAL 13.5 AFR.  It is MY bet...  you will be darn close.

[mayor]

I think for the test to have merit, the test parameters should change.  If I understand correctly, the end goal is for a DIY closed course tuner to be able to tune the wide open based on trying to get their wide open throttle on board sensor voltage to within a particular range.  If that's the goal then the actual test should center around finding DIY guys that will try to achieve a predetermined o2 sensor voltage at wide open, then they go have their bike tested on a dyno to verify the wide open afr to see if the wide open is in a predetermined safe afr range.  Maybe I'm getting ahead of the data, have we reached a consensus yet on what the voltages at wide open should be?  Is the desired voltage different between lambda and afr based bikes? 

does using the narrow bands out of their published range slow down the rate at which they sampled accurately?  what sample rate is desired when using the datamaster program to sample generic o2 data?  Has there been a formula worked out to tell a DIYer how much to adjust the ve's based on a certain voltage range off of the desired?

no worries Fire, was just trying to let Bob know I was over here and was willing to join in the discussion again.     if you look at my e-mail addy, you'll see I'm not trying to hide who I am from the admins. I also used the same avatar and screen name, so that those that may frequent other sites would know who I was as well.  I have another log into for this site from years ago, but I thought being incognito would seem shady once some of these guys eventually figured out who I was.   

[Buckeye_Tuning]

wikipedia says the range of the NBs is what I stated in my previous post.  .9 lambda to 1.1 lambda.

I do NOT think the test should change Mike... you tend to get ahead of things because you think ahead of things.  The FIRST order of business would be to simply check the accuracy of the NBs.  THEN and only then, could we move forward intelligently with that in hand and discuss ways this could be used.  Why do a beat down on a bike to simply check if the NBs can accurately read 13.5AFR?  Why use a dyno?  One could sue a twin scan and a TTS and achieve these goals I have set out.

[sadunbar]

If after 261 posts, no answers to the original questions raised, and ending with the thread being locked down, is there really anything to gain by starting the same conversation all over again with primarily the same participants?  Just wondering....   

[Fired00d]

....

no worries Fire, was just trying to let Bob know I was over here and was willing to join in the discussion again.     if you look at my e-mail addy, you'll see I'm not trying to hide who I am from the admins. I also used the same avatar and screen name, so that those that may frequent other sites would know who I was as well.  I have another log into for this site from years ago, but I thought being incognito would seem shady once some of these guys eventually figured out who I was.   

Have no problem w/you joining the discussion (or the site) as all are welcome. As you well know from being involved w/other site(s), personalities and agenda's are something we (as administrator's) have to deal w/on a daily basis. As you would, we would like for whatever happens on another site(s) not to bleed over and come to this site (we each have enough to deal w/on our individual sites ). Since you mentioned you had another log on from years ago I would appreciate it if you would delete it (or PM w/what it is/was so it can be removed) as we don't allow multiple accounts from users (as you can well understand why). Thanks.

 
Ride Safe,
Fired00d
 

[mayor]

holy cow wurk, when did you sneak the book in. 

I do NOT think it would take away any need from pro tuning.  The same guys that go to pro tuning would still go... there's a LOT more to it than getting VEs dialed in.  But...  I DO think this could help the members that will DIY their bikes, regardless.  The idea has been floated AND worked upon to make the NB sensors in our bikes used for 80+ kpa and would eliminate the need for 'extending' VE tables.

I have found, on my very own, that when using calibrations that are AFR cals... extending the VEs into the 90-100 kpa ranges works well.  ....But...  I have also found the same to NOT be true when working with Lambda calibrations.  Sometimes I find those to be lean.

I agree, and have has similar success with the afr based narrow bands.  The lambda doesn't seem to extend to well based on what I have seen either. 

A LOT of folks have stated that the NB can NOT reliably measure 13.5 AFRs.  I disagree with those guys.  Some have made good arguments on why they will NOT work, but some of those ideas have holes in them.  For example, ... "The OEMs do not use them for fuel control, so we can't either."  Know what?  I agree with this statement 100%!!!!!  I do not think we could.  But.......

I am NOT proposing, and neither is Bob that we use these for fuel control at 13.5 AFR.  Nope!  Instead, all we would need is for a Steady state tuning run to show a single 13.5 AFR=13.5 AFR connection to things.  These sensors may not be accurate in 13.5 as fuel control devices that warble fuel rich, lean, rich, lean 8-50 times a second...  but what is to say those self same sensors wont read a bike that is held steady at a certain Map and RPM to gather data with?  I think NOTHING says that.

I don't disagree, I think there is a difference between using for fuel control and using it to take a reading.  That's why I was curiouse why this hasn't proceeded to the next step (DIY testing). I have no doubt that there is enough data already collected, so any more time collecting is just slowing down the inevitable DIY test.

[mayor]

I propose doing this in an easy fashion on a bike, easy on the bike, also.  Set up a bike that outputs 13.5 AFR at say 3000 rpms and 60 map.  Use Wide Bands to do this.  This will not beat on a bike like WOT would, it's just a test after all.  Once the bike outputs like you/we wish, look at datamaster and see what millivolt reading is showing.  Simple!  It does NOT matter if we do this test at WOT or just above idle.  The test is all about the accuracy of the sensors.

the problem with trying it at 60 kPa is trying to get a good stable 60 kpa.  If I was reading Bob's statements correctly, the only way to accurately test this is for the conditions to be stable as possible. The wide open throttle conditions are actually quite stable, and probably failry predictable.  There is no way to get a good stable 60 kPa reading on the street, and I'm not sure how stable that is on a dyno either (although, I guess it would be possible)....not to mention that EGT would likely come into play at a lower value than what the test was being designed for (wide open).   

Fire, I think my previous log in was quaker state or something there abouts.  I'm from PA, so that made sense at the time. 

[mayor]

If after 261 posts, no answers to the original questions raised.... is there really anything to gain by starting the same conversation all over again with primarily the same participants?  Just wondering....   

I agree with Tsani, this forum opens the discussion up to some new folks that haven't yet weighed in on the subject.  There was a guy a while back that was very vocal on the tuning range limitations of narrowbands, and I'm thinking he is an active member of this site so maybe he'll see the thread and jump in and let us know if he has the same opinion now.

[cvofbme]

Actually, its not really the same participants ......,

nor for that matter is it the same question.

[mayor]

good point, and I think this question is much more to the point with a specific goal in mind...can't really say I felt that way for the other three threads I read on the subject.

[hrdtail78]

http://www.rbracing-rsr.com/rsrgauge.htm

This guy has been selling and doing just that for years.  Back in 2005 I bought his setup and road tuned my 128 bastard build hard tail.  It still runs like a top.  Single wire, narrow band sensor.

BTW  Clost to the bottom of the page there is a great graph that explains how pressure really affect broad band sensors.  Shows less than 1 BAR also.  Just incase you believe that a HD exhaust might not always have atleast 1 BAR.

[Steve Cole]

The O2 sensor does nothing other than read the difference in the free O2 in the exhaust from the air surrounding the pipe, that's it! It does not matter what type O2 sensor it is, Wide Band, Broad Band or Narrow Band they all do the same thing. There is NOT ONE O2 sensor that reads fuel mixture, period. It is all inferred from the difference in O2 measured. So what is different is the accuracy of what they are doing, NOT what they are doing. Since NONE and I mean none of the systems currently being used are accurate to begin with it all boils down to what is good enough for the task at hand. Once you establish what is good enough, you can then see what will do the job.

The truth is most of the Broad Band based systems on the market sample the exhaust at something, at, or less than 10 HZ, that's 10 sample per second or less. With an engine operating at 1000 RPM each cylinder is firing at 8.33 HZ So today's Broad Band based units handle that just fine but lets move on to 6000 RPM which moves the engine firing to 50 HZ and there is the poor old Broad Band running at 10 HZ or less still. Now what are your really measuring and displaying as it sure isn't what each firing cycle is doing! The stock ECM samples the factory O2 at engine speed so it is running at 8.33 HZ @ 1000 RPM and 50 HZ @ 6000 RPM.

So which system to start with do you think has a better chance of giving you a more accurate reading?

Now everyone ASSUMES that a Broad Band is more accurate than a Narrow Band Sensor and if you follow the manufacture specifications it is. Only problem is, I have been unable to find ANY of the current units on the market today that follow the O2 manufacture specifications! So now we've got Broad Band units that donot follow the manufacture specification for the sensor and you want to rate that against the Narrow Band sensor.

So the real question becomes can you use a out of specification Broad Band unit to test against a Narrow Band sensor thats used per the manufactures specifications? Then, can you expect those test to be accurate enough?

So it's like trying to measure two parts, one that is 0.0002" bigger than the first part with a tape measure, the results will not be worth a chit to begin with. The truth is a HD engine really doesn't much care about the AFR at WOT as long as it's in a safe range. That safe range is anywhere between 12.5 - 13.5 to 1. If you run a properly tuned HD at 12.5:1 AFR and then again retuned properly at 13.5:1 the power output may change by about 2-3 HP and 2-3 ft lbs! Less than most any of us will ever feel!

So with this knowledge, can a Narrow Band Sensor be good enough to tune with, like a Broad Band Sensor is thought to be?

Can you measure the voltage output from a Narrow Band Sensor and read it accurately ENOUGH to tune with?

Will the voltage from the Narrow Band sensor be stable enough to do the job at hand?

You got to answer these questions to truely understand what you are trying to get too. So if you are going to try and compare a Narrow Band sensor reading against a known out of specification Broad Band your going to need to accurately record the conditions so the Broad Band readings can be properly corrected if you really want to know the answer OR you can determine if the output of the Narrow Band sensor is just as stable as the Broad Band unit is in the same conditions. If they are then the answer is simple.