HarleyTechTalk

Quote from: whittlebeast on October 30, 2011, 09:12:57 AM
I just swapped back to the US revision ECU.  Problem solved.  It has to do with the way the Euro interprets the lack of a o2 signal as opposed to the US code.  Testing and looking at the data is everything in this sport.

Might be more along the lines of the way the Euro emissions works.  Does their exhaust plumbing have some active components that ours don't?  Problem was likely solved in this case by removing the cause of the problem: mixing/matching parts ad hoc.

Quote from: whittlebeast on October 30, 2011, 09:52:50 AM
I stand by my VE definition as a great functional definition.  It is not technically correct, but I really do not care.

You might not care about correctness, but it'd behove you to give some consideration to those who are coming to you with questions in a process of trying to learn how things work.  First give them the facts/truth, then once they grok that, maybe only then throw some of your idiosyncratic notions their way once they've got the basics well understood?

I've also wondered about the comments the bike runs like crap.  Following all of these Sportster related posts and how poorly some think they run.

My bike is basically the same thing Andy has, 07 1200 Low all stock except I've hogged out the mufflers a bit.  I think Andys is all stock except the air filter.

I've always thought my bike ran very nicely, it gets fair gas mileage and I've never noticed any flat spots anywhere, but then again I don't rod on the bike too much.

I did a TTS dyno pull with it once and followed Steves instructions for best use of that.  3rd gear, 1500 RPM, roll on WOT and go til it hits the rev limiter.   Well it took off and there was no hesitation, and when I hit 95 mph and the rev limiter still did not come in, I backed off the throttle. 

When I got home and looked at the data log and framed the dyno pull around the start of the WOT and the end of it, the vehicle speed was 98 mph and the dyno pull showed 67 HP and 71 Ft Lb TQ.  The time elapsed was 17 seconds, now those numbers might not be quite what a race machine would do, but it did that pull with absolutely no problems.

Blackcherry

Email me a data log of about an hour ride some time.  I would love to snoop around at how you have the tune set up.

Beast

Sporty, I've thrown together a small example of how calibration table cells get used.  It should illustrate how changing one value can have ramifications that cover a broader area than you might've considered.  It should also serve to illustrate why it's so important to get the VE tables, especially, populated as accurately as possible since they are the foundation for so many calculations.

Looking at some (Big Twin) datalogs, I see the TPS records values to the tenths of a percent and RPM values are to whole numbers.  VE values are to a tenth as well, so I'll round them for this example:  say you're at 27.2% TPS, 2281 RPM; what is the VE value the ECM uses for that fuel calculation? 

From the table segment attached below, that resultant VE would be "91.9".

Do you see how arbitrarily changing a VE table value can affect other areas?  What if I'd changed one of those four cells by a few digits just because I'd "seen a hole" in a log being viewed in a graph?  What if the data from that timeframe didn't actually all perfectly coincide with what was really going on then?  What are the chances it would actually be the correct thing to do?

Conversely, if you wanted to change just the VE value for 2281 RPM @ 27.2% throttle, you'd have to change four cells appropriately, and that would affect an even wider range of operation!

Take this into consideration when you see an RPM, TPS, VE, and VE-new value.  Both the VE and VE-new values will most every time be the result of proportional averaging, so merely taking the VE-new figure and sticking that somewhere into the VE table just ain't the right thing to do.  What's more, you don't know where the AFV cells are located in relation to the affected VE cells, and those AFV cells, proportioned, along with current corrective values that aren't (yet?) recorded in the AFV tables are what goes to produce that "VE new" figure.

I'm not trying to scare you off, just trying to get you to slow down a bit.

[edit: I removed the first image and attached the updated version.  For some reason I'd started incrementing the RPM by 25 and in the middle switched to 50, so all the numbers apart from the top and bottom rows were incorrect.  All better now...]

[attachment removed after 60 days by system]

Quote from: whittlebeast on October 30, 2011, 12:19:46 PM
Blackcherry

Email me a data log of about an hour ride some time.  I would love to snoop around at how you have the tune set up.

Beast

Andy, I won't get anymore riding in this season, but in the spring I can get a longer run that I can shoot off to you.  The tune I have is something that MOCO created with the SERT back in 08.  I hit them up hard about the cold start stumble and after they (dealer as instructed by MOCO) had reringed the engine and changed all of the EFI components to no avail.  The MOCO took a new ECM and a SERT, tweaked the warmup mode (I'm assuming) and overnighted it to the dealer and the stumble was cured. 

I don't know exactly what they did but I imagine they have the ability to load a basically stock map into the ECM with the SERT and then adjusted the warmup.  One thing they missed was the rev limit, on my dyno pull the data log showed 6591 RPM and the limter had not come in yet.  So this makes me wonder if they maybe used a stage 1 map and readjusted it to a stock configuration or something like that.

FWIW, I'd like to thank Glen for offering his knowledge and expertise to this forum.

What folks like Glen do is independently put credability to a few who have spent their free time helping members with EFI running/tuning issues.
Of course most are bias to what they feel work best for them based on their own experiences.
Thanks again!
Bob

Yes FLTRI, a very good thread, great contributors and it is interesting.
You guys have posted so many questions for me it is going to take a while to sort them out and answer them.
With a fair amount of snow on the ground, intermittent power outages, a strong northeast wind and other stuff in life the bike is grounded here for a while.
Glens, when I first got the Sportster it was the biggest disappointment power wise, reminded me of an observed trials bike with a bad tune and a bad suspension. Now, just a whole different ride, but the VE's are not optimized yet, the first step in refining a decent tune.

Quote from: whittlebeast on October 30, 2011, 09:52:50 AM
I stand by my VE definition as a great functional definition.  It is not technically correct, but I really do not care.

Beast

Shame you really don't care about being technically correct since there has to be a common basis for discussion and exchange of information correctly. So if you are talking about apples and the thread is about oranges, it really shouldn't matter since one could just say it's fruit.

Since it is just fruit:
Volumetric efficiency in internal combustion engine design refers to the efficiency with which the engine can move the charge into and out of the cylinders. More specifically, volumetric efficiency is a ratio (or percentage) of what quantity of fuel and air actually enters the cylinder during induction to the actual capacity of the cylinder under static conditions. Therefore, those engines that can create higher induction manifold pressures - above ambient - will have efficiencies greater than 100%. Volumetric efficiencies can be improved in a number of ways, but most notably the size of the valve openings compared to the volume of the cylinder and streamlining the ports. Engines with higher volumetric efficiency will generally be able to run at higher speeds (commonly measured in RPM) and produce more overall power due to less parasitic power loss moving air in and out of the engine.

Now that is a banana I can understand. And if you ask me it represents a fairly fixed number to which all else is determined around. Doesn't sound like a fudge factor to me. Get the VE's set right and everything else is based off of it. I do understand your point about using it as a "fudge" factor to make up for something else you cannot seem to correct otherwise for what ever reason, but that fact does not change what the VE really represents.

Quote from: FLTRI on October 30, 2011, 01:20:19 PM
FWIW, I'd like to thank Glen for offering his knowledge and expertise to this forum.

Why, shucks Bob, you're more than welcome.  Really, you're the one needing thanks.

QuoteWhat folks like Glen do is independently put credability to a few who have spent their free time helping members with EFI running/tuning issues.

Why does that sound like I'm not doing this on my free time?  Who needs to know where to send the check?  Hahaha!

Quote from: glens on October 30, 2011, 10:02:45 AM

Quote from: whittlebeast on October 30, 2011, 09:52:50 AM
I stand by my VE definition as a great functional definition.  It is not technically correct, but I really do not care.

You might not care about correctness, but it'd behove you to give some consideration to those who are coming to you with questions in a process of trying to learn how things work.  First give them the facts/truth, then once they grok that, maybe only then throw some of your idiosyncratic notions their way once they've got the basics well understood?

I agree with Glen and Tsani. The problem comes in when folks don't know that they way they are being taught is not by the book correct, then they get this false sense that when others disagree with the tuning tactics it's because those others are not smart enough to see it from a different angle.  You may not care that your definition is technically right, but the Delphi system cares.  By adjusting the wrong tables to achive the desired results, then all other tables become wrong and do not function correctly.   

Quote from: Sporty 48 on October 30, 2011, 01:56:05 PM
Glens, when I first got the Sportster it was the biggest disappointment power wise, reminded me of an observed trials bike with a bad tune and a bad suspension. Now, just a whole different ride, but the VE's are not optimized yet, the first step in refining a decent tune.

I think you are missing the point entirely.  What we are trying to tell you is what you think is optimizing the ve's may be nothing more than you tinkering with the numbers and getting a false felling of improvement.  You might even be getting improvement (meaning a feeling of more power), but that could be because you are no longer running in a fuel effecient afr range.   

Here is a screen shot comparison of how I set up a 150 HP rice bike compared to a 70 HP Harley.  These are actual AFRs coming off the same wideband.



Have fun tuning

Beast

What follows is an excerpt of the thread on another site, mentioned earlier in this one.  Each post is preceded by a link to that post, so I guess one could merely jump into the thread at the first link, but there are a lot of adverts in the pages, and you'd need to flip through several of them (fun enough in its own right, I guess), so I cut-n-pasted pertinent text and included imagery as "necessary".

Andy, are you happy and proud?

http://xlforum.net/vbportal/forums/showpost.php?p=3596781&postcount=577
KevinJ:
"So I just did a few log runs....
This is what it looks like: horrible!


http://xlforum.net/vbportal/forums/showpost.php?p=3596789&postcount=579
whittlebeast:
"WOW that thing needs tuning....

Is the intake and exhaust stock?"

http://xlforum.net/vbportal/forums/showpost.php?p=3596790&postcount=580
KevinJ:
"Beast, yes, that is the stock Harley tune... And the whole bike is stock, stock airfilter, stock exhaust (with cat convertor)

edit: I did remove the flapper valve in the airbox. Don't know it that would change much."

http://xlforum.net/vbportal/forums/showpost.php?p=3596797&postcount=582
whittlebeast:
"If you want I can set up your tune to get started..."

http://xlforum.net/vbportal/forums/showpost.php?p=3598346&postcount=589
KevinJ:
"This is where I'm at with the first custom map (Thanks Whittlebeast)

So now I've added 3% to the red dots and removed 3%from the blue dots.
And I also made the rear 4% richer as of 3500rpm from 0 to 10 % TP (Like Whittlebeast did with the front yesterday)"

http://xlforum.net/vbportal/forums/showpost.php?p=3600032&postcount=599
KevinJ:
"This is where I was yesterday

And this is where I'm at today. I'm getting there...
"

http://xlforum.net/vbportal/forums/showpost.php?p=3600050&postcount=600
whittlebeast:
"Just keep picking away at it 3% at a time, It will take a quite few rides before you will get to the point that you are just chasing yourself in circles and it is time to move to step 2. You will get it."

http://xlforum.net/vbportal/forums/showpost.php?p=3601685&postcount=605
KevinJ:
"5th map revision and this is where I'm at:

I do have a strange starting issue since I got to map 4: when the engine is hot and shut off for a few minutes (fuel stop), when I want to start again, it hardly wants to idle and even drops below 500rpm sputtering along before it dies.
I did add fuel to the 0% TP at 1000 and 1250 rpm to get rid of those lean spots. So now I reverted those settings back to the (lean) "map3" settings. Hope it'll work.
Oh, and the plugs are really black now.
But I did drive more than 500km this week, making logs "

http://xlforum.net/vbportal/forums/showpost.php?p=3601719&postcount=606
whittlebeast:
"Plugs should never be black on an EFI bike. Something is messed up."

http://xlforum.net/vbportal/forums/showpost.php?p=3601777&postcount=607
KevinJ:
"Half of the map is still messed up, so I guess it might have something to do with the bike being rich at idle and it idles about 5 minutes while I try to get it in the 2nd garage.
And the strange idle behaviour after a short shut-off probably doesn't help (because it really smells like gas then)"

We found what happened to his tune.  He PMed me the logs and I found it in about 30 sec.

I think I missed the point of your post???

Beast

I highly doubt you actually missed the point, Andy.

It's part of a recent trend, here on HTT, to sayy that there is only ONE method to tune by.  All other methods are suspect and not to be used.  I, personally, kind of, take offense at the shots made at a tuner who uses 5 gases and tunes using  both AFR and VEs to dial in his tunes.  That particular tuner advocates open loop only tuning, and states this when going to him for a tune.  99&44/100% of folks that go to that tuner... or ANY tuner for that matter, do NOT attempt to alter their tunes after the fact.

YOU, Andy, have tried to make a member here alter an unalterable tune, and you need the new HTT punishment for that action.  You were NOT correct in your approach to this member and should have INSISTED that member load up a base tune to start from instead of trying to play with a tune, of which NEITHER of you know about.  OK, fine.

YOU, Andy, have corrupted this member's brain to the point he cannot see reality.  Truly, you did.    I feel this member either will or will not get straightened out.  But........  now YOUR punishment is to be shown your total lack of competence by using posts from ANOTHER forum to show a 'tune' that went awry ..............  and obviously...... that MUST be your fault.

Watch closely here.............  you are next for banishment.  Why?  Because you do not toe the party line, is why.

Instead of the free flowing of ideas... some are old arguments, some can be new and novel...  this section has turned from the AFR section to the..............  one MUST think the same as 'us' or you will 'get it for sure' section.

Andy, rest assured that I believe in the 'normal' (around here) way to tune and set up maps....  but it IS beyond me why some folks here think those methods are the ONLY methods, and also that 'dissenters' need to be quieted.
BVBob and Stroker have left the building........  it seems to be ALL about ONE method of tuning and ONE way to use that method.  Open loop by BEAN, etc is thought of as 'dumb' here anymore.  Nothing is equal and there IS no free flow of ideas, etc.  You need to watch yourself.

As much as I love this site...  it is easy to see that we are all NOT equal on here.

In the case of that tune It was traced down to a typo where the owner multiplied a series of cells by 1.3 instead of 1.03

Beast

Andy, there's a lot of dark blue in that first log he did using that "starter" calibration you'd evidently provided.  Was the "1.30" vs. "1.03" committed that early?  When was it committed?  Everything up until his "starting/idling" statement looked okay to you, didn't it?  Or did you even look?

My point is that your method isn't bad because it's novel, nor that it departs from the "standard" way.  It's bad because you advocate it to folks who don't have a clue and because you evidently don't do anything to teach them even correct terms.  And because you, as was stated a bit earlier, advocate using it against a non-"conventional" method of tuning which is not readily receptive to such methods.  Furthermore, because in doing it you rely solely on data which is nowhere near complete enough in most cases, because you manipulate that data, and on top of that, it's currently overly replicated when using that other kit.

If a person were able to hold primarily the centers of all the cells in the VE tables, each at steady-state operation, there's not too much to say against the method itself; at least in respect to the limited aspects it addresses.  I haven't noticed any evidence that you advocate that.  Or that you advocate avoiding transient conditions as much as possible.

I'm somewhat confused by your advocating CLBs of 450, too, as well as holding stoich or richer on overrun.  In fact, there are several aspects of what you advocate that confuse me.  But that's all right, because I'm not going to play with things that way so it doesn't matter.

But I think you're flirting with disaster, Andy.  Other people's disaster.  You (evidently) hadn't even asked that poor soul what sort of modifications he'd performed, nor suggested a suitable starting calibration for them.

I hope he didn't wash down his cylinders too much before it'd got caught.

It would also be beneficial to keep such help either totally off the boards or totally on them.  That would keep things a bit safer for others.

Wurk:  I don't much understand your post.  Sorry.  Some of it I think I agree with, but some if it I don't.  What I don't like about what Andy's doing is that he admittedly doesn't care about even using correct terminology, especially with rank neophytes, and that he doesn't bother to (much) enter into any meaningful discussion with those who have at least a clue.  That scares me.  He portrays himself as a vast resource of knowledge and experience but does precious little to back it up.

Since I feel some of your derogatory remarks are pointed towards me, I'd like to ask you: What sort of notion do you have as to what my idea of an ideal tune is, and what method of deriving it?  I think most people who bother to read what I write would see that as a rhetorical question because I don't advocate anything in that regard.  Am I a fan of closed-loop operation?  You betcha!  Do I like to keep it simple by sticking with only the Delphi components for daily operation?  Yes sir!  But that's about as far as I go with it.  I'm pretty sure about that just now, at any rate.

This thread is about doing what Andy does, and more than tangentially, how he does it.  I feel bringing that other information over here is both entirely pertinent and warranted.

As far as those other two guys you mentioned, I guess I've missed where they've discussed in detail why they advocate that which they do.  I'd love to see it.  About all I can think of along those lines are merely statements made on their parts, nothing further.

Whoa Nelly, easy boys.
Wurk Truk, pretty funny, sarcasm is best applied sparingly, but you have pulled it off elegantly with a sledge hammer.
There does seem to be a level of intelectual rigidity and a rather flaccid response to new ideas here at times, therein lies the challenge.
Sharpening the logic requires work and rigirous thought. We have that here.
Now as for Andy, helping others, simplifying complex terms and in general having fun while tuning, go dude!
Many times Andy Whittle Beast has cautioned me when considering performance options and he has refocused me back to the critical elements of the tune: VE's, AFR and Timing.
Did I miss anything?
I can't wait to ease into closed loop tuning.

Quote from: wurk_truk on October 31, 2011, 11:19:24 AM
I, personally, kind of, take offense at the shots made at a tuner who uses 5 gases and tunes using  both AFR and VEs to dial in his tunes.  That particular tuner advocates open loop only tuning, and states this when going to him for a tune.  99&44/100% of folks that go to that tuner... or ANY tuner for that matter, do NOT attempt to alter their tunes after the fact.

If those comments are directed towards my comments about the ve's needing to be correct, for the other tables to work right...you are reading too much into what I'm saying.  First, I have never met Bean so I have no personal reason to say this- I would trust Bean to do a tune for me.  That's saying a lot, because I can't say that the list of folks that I would trust is very long.  Any comments that I made was not to discredit Beans tune.  If what he did was dial in both afr and the ve's to get certain numbers, that's irrelevant to anything until someone starts messing with things.  This is likely going to be an issue as soon as someone who doesn't realize how the things were set starts adjusting things at face value. If the tune is never touched, no biggie.  I don't personally have an issue with open loop tunes, but then again I also like closed loop tunes.

Second, so you find it offensive when some would question how the afr was derived...but you are fine with a fellow starting multiple threads talking how he adjusting the lean spots out of an open loop tune that was done by Bean?  I personally would trust Beans equipment over those cheap assed narrow bands, and I have a hard time believing that a open loop tune set to 14.2 or so will have lean spots that are lean enough to be felt by the operator. I have to say, I find the very thought absurd.   Heck, I had lean spots in 15+ afr range on my carb bike that I never felt.  I just can't see a tune that Bean did drifting greater that a full afr.  I think the micro tweeking to fix lean spots discusion does far more to discredit Beans tune than us discussing the methods of how he reached his verified afr's.

 

My method merely preaches to look at the data.  Correctly installed o2s a wicked fast and wicked accurate at 14.6 and are totally trusted in every closed loop tune.  When doing a closed loop tune and aiming for 14.0, the o2 should almost never swing to the lean side of 14.6 as that is over a 1/2 AFR from your target AFR.  If 90% of the o2 hits in one area are lean, you have a hole.  If that hole only shows at some weather condition that Bean never saw in his lab on the day he tuned the bike, it it not is fault.  Just a weather condition he never ran into on that day.  My bet is Sporty48 never messed with 90% of the VEs as Bean had them set.  I also bet he never touched any of the timing curves that Bean set.  AE, DE, you name it is all as Bean set it.  It was just the few lean holes that always seem to show up on the data logs.  These are the only VEs that messed with and fuel was only added.

I wish Sportsters could run closed loop but the AFR swings are simply to huge.  These bike are just to sensitive to having that much fuel pulled out.  If it was 3% pulled for a couple of tenths of a sec, it would be fine.  We are forced to accept  open loop and shooting for 14.0 and accepting the fuel and weather changes.  I wish we had access to the o2 settings but we don't at this point.

Keep in mind that most all of the PC3 and PC5 tuned rice bikes running around are all running open loop and are making insane HP.

Beast

If 90% of the hits are on the lean swing, all that means is that 90% of the hits that were captured were on the lean swing.  It doesn't mean that any more than half the swings the engine uses were lean.  Like you said, the sensors are wicked fast.  So is the ECM in driving them.  It's just not wicked fast in spitting data out onto the bus, since that area of its operation is relatively inconsequential and the bus has more important things to do in normal operation.  Not to mention the bus type is relatively antiquated and not necessarily even up to the task of passing all that much data.  Are you positively sure that the sportster ECM code makes bigger swings than the big twins?  Do their smaller injectors have anything to do with their apparently larger fueling swings in the data logs?

You are aware, aren't you, that Bean uses the VE tables somewhat conventionally along with the AFR tables somewhat unconventionally to generate the tune, right?  By altering any of his AFR commands you've changed things in the VE tables as well.  I'm glad you think you can do better with a smidgen of data off the bus during uncontrolled operation as compared to his multi-gas analyzation during controlled operation.  It shows you can think outside the box.  But when folks who've been working intimately with the box for years, much more intimately, try to tell you what the box is capable of and what it's not, you'd do well to listen to them instead of trying to do what you likely won't admit publicly you're really trying to do.

I'm sure Bean does what he does because the two table types don't have perfect coincidence.  Heck, they're not even using the same two types of information as each other in the sportsters.  Which brings up another point.  The VE tables, when using Alpha-N axes, cause each cell to cover multiple MAP values, which means multiple airflow conditions.  Are you sure that your system of tuning properly takes this into consideration?  Isn't it rather using hit-or-miss squared?

If the sportsters used individual intake runners and spun to twelve grand, using double the valves per cylinder they now use, I'd bet they'd make wickeder horsepower too, even running closed-loop over much of the range.

I personally use both Alpha-N and Speed Density logs to look at data.  On the Yamaha, I use Speed Density to make the vast majority of the VE corrections but then I always look at the VE trace to get to holes that are more throttle position based.

Beast

There is essentially four ways to tune a bike that I am aware of.

1) Tune the majority of the tune utilizing the NB 02s to develop the VE News.  Then you use the average or the NE News (after applying filters) to develop the the VEs.  You then extrapolate the VEs beyond the areas of the map that you are willing to target the valid range of the NB O2s.

2) Use target AFRs that are not in the range of a standard NB so you use a wideband O2 and lean on the VEs or (VE correction factor) to get the actual AFR to hit the target  AFR.  This is essentially how the guys that can not hack the ECU are stuck to tuning.

3) Use a dyno and a 4 or five gas and for the most part shooting for a near constant CO of about 4-6%

4) Set the TPS to a position, have the dyno hold the RPM and then tweak the AFR/VEs and timing to maximize HP/torque at every combination of TPS and RPM that you can get to.

Have you ever looked at the VE tables after the same bike went thru all 4 methods?  Does one method excel in drivability in all situations?

I do see some amazing differences (and clues) looking at lots of different maps.  I hope some day we can get this site to debate this sort of thing.

Quote from: glens on November 01, 2011, 04:53:06 AM
If 90% of the hits are on the lean swing, all that means is that 90% of the hits that were captured were on the lean swing.  It doesn't mean that any more than half the swings the engine uses were lean.

This is a point that interests me;
On my twincam the PV logs show another interesting tendency, when the clb's are at 700 mV, the sensor voltages average (count or integral) about 840 mV, and it looks like the sensors are doing crossovers  about once every 5th second.
With clb at around 500, it looks like the sensors are doing crossovers about once every second (like they are supposed to do). Until spring and further testing I can only speculate around several possibilities:
-Are the NB sensors that much slower at a higher clb?
-Is the ECM dataoutput the reason? Why is there a systematic skew one way or the other?
-Is the datacollecting on the PV the problem? Again, why is there a systematic skew?

What do people using TTS see when they log sensor voltage at higher clb's, what is the mean time between crossovers compared to at about 450 mV?

Andy,
Not sure why you accept that nothing you can do with any software that will change the voltage output of a sensor.

No matter what you set CLB's the sensor puts out exactly the same voltage as with no CLB.

Now the software takes those O2 sensors' output voltages and may apply an offset to tell the software it's running leaner/richer than targets...then adds more/less fuel to get the exhaust output the target/desired AFR.

I have to note, it appears you make statements simply for your amusement listening to those who actually know why your statements are incorrect...then you simply change your posts kinda like keeping folks wondering which cup you put the pea under.

Remember Andy, you are trying to compete/educate some folks here who either designed the calibration tuning software or use the software/hardware for emissions testing which cannot be done without proper testing using really high speed/reliable equipment used for resolving emissions and/or rideability issues.

What do they use for this precise measurement? MLV? Datamaster?, NB? BB? O2 sensor?, or any other aftermarket software/hardware? NO!

I believe what some of these experts are trying to get you to see and understand is the equipment and software you are using to determine what is "actually" happening WHEN it is actually happening.

Can you make an educated guess and have a relatively good chance logging over long periods will show some data that couldn't be expected reliable with the sample rates included in the equipment we all use to tune bikes.

These few talented calibration guys provide the luxury of having them to educate us.
They use sophisticated OEM calibration hardware and software to which we have no access (due to costs) to resolve internally hardcoded software and hardware that affects running characteristics.

What I would do if I were you would be to ask for advice and take it...unless of course you can effectively discredit the data with advice you've received.

There is a lot of information you could use to help understand what you are really doing and advising?..for example how much run quality and power change is realised with a 3% fueling change. ie: 14.2 vs 13.8.

Hoping this is taken in the spirit in which it was intended,
Bob