How to get BEST 'Home Made Tune?

[FLTRI]

Glens,
All good info, thanks for sharing. Hopefully those who are under some O2 (broadband as well as switching) sensor misconceptions will learn from it, which is readily verifiable. That should help in future dicussions.
Bob

[Steve Cole]

Glens hit the nail on the head this time pretty good. I have one small correction to add. There are THREE types of sensors currently on the market, Wide Band, Broad Band and Switching Sensors. Wide Band sensors are not used by any of the Aftermarket units today but they like to play on the words and call Broad Band sensors Wide band, since they will measure a broader range than a switching sensor under some very controlled conditions. The aftermarket has conveniently left out the part about "controlled conditions"!  So if your replacement sensor run about $100 all you have is a Broad Band sensor and if you want real measurements then you need to control the conditions. A true Wide band sensor cost about $900 per sensor. They work over a large range and do not have the issues Broad Band sensors do. Broad Band and switching sensors are made from the same base technology. The Broad Band sensor was made to check Catalytic Converters in the Automobile industry and are not used for fuel control. If they were any where close to as good as some of the aftermarket companies would want you to believe do you not think the OEM's would use them for fuel control?

Truth is they were made as a cheap way to test the converters for the OBD II requirements and they do that very nicely. The aftermarket has taken them and tried like hell to make them work where the manufacture says they will not.

[Don D]

Can you explain the differences between a person inputting the VEs from a TS2, etc... and a v-tune's VEs?  Please

Assuming you have TTS the adjustments from the TS will manually need to be inputed then commited by uploading to the ECU. In TTS VTune the software changes are made for you and you just commit them.

[glens]

You can't just put in whatever CLBs you want after open-loop tuning if you care about what your only-open-loop areas will be doing afterward.  Why would anybody tell you what you're saying you were told?

When you're v-tuning, whatever your CLBs are is what the ECM uses to say "with the displacement and injector sizes specified, it takes this much VE to achieve 14.6:1 AFR here".  If you stick VE values in which were otherwise derived to achieve an actual 14.6 and are using higher CLBs, more fuel will be required.  This gets picked up by the adaptive fueling and wherever this coincides with an AFV cell, the AFV gets modified to reflect the "necessary" change.  The AFVs effectively act as VE modifiers against what's in the programming.  They get applied at all times, either open- or closed-loop.  The last AFV cell in an "outer" area gets used everywhere else the rest of the way out.  If it gets modified from 100%, it'll effectively be altering all the "outer reaches" VE values as well, pulling the AFR down from what's in the table there.

If you v-tune with the CLBs where you want them to be while in normal service, the extra fuel will be applied directly via the VE tables and the AFVs will remain closer to 100%.  You can then get all the outer reaches of the VE tables set via a TwinScan and they will remain right where you placed them.

So in the end the results wind up not being quite the same; there's a distinction.  If they're achieved with v-tuning, the AFVs will be ~100%.  If they're achieved with "actual" VEs from an exterior source but using the same CLBs as the v-tune would've used the AFVs will be ~103% (or whatever).  AFVs get applied everywhere they can so you need to take this into consideration and tune the outlying areas afterward with as-close-to 100% AFVs as possible.  It's a shame we can't know where the AFV cells are or what they contain.

I've been thinking about something I touched on earlier.  There's nothing in the TwinScan manual about adjusting the readings for different fuel types/formulations, so does it always indicate 14.6/14.7 at stoichiometric burn, regardless the actual AFR that occurs at?  If so, this would be because the controller is calibrated against/as-if-it-were "pure" gasoline, just like the pre-lambda Delphi calibrations.  And presumably the two ECU replacement systems, too.

So with any of them, if you had stoichiometric burn with pure ethanol, it would indicate as "14.6/14.7:1" AFR, even though it actually took 9.0:1?!  Can this be so?  Certainly, if the manufacturers were aware that stoich would occur at ~14.2:1 with E10, and these controllers were calibrated to show purely actual AFR (but they really couldn't be if you think about it; not without knowing what fuel they were working with if not "straight" gasoline!)...  I lost my train of thought.  Okay, look at it this way: at a stoichiometric burn of whatever fuel, there will be no "pumping" going on, either towards the sensor element or away from it.  Without "knowing" which actual AFR this is based on (fuel choice), it could only tell you "it's lambda equals ONE".  This is relatively meaningless anymore when the ECM calibrations use hard numbers like "14.6".  In reality, the ECMs are using lambda behind the scenes but the interface indicates lambda=1 as "14.6".  I guess what I'm getting at is that all these AFR numbers are to be used as if (pure) gasoline is the fuel and likely-as-not these days don't have much bearing on reality.

The reason I went off on the sensors earlier was because (wasn't it?) you, WT, who went off on them in the other direction.

Steve, I see I didn't state the distinction, but I did make it when the only place I use the term "wide-band" was when I quoted it just like that at the start.  Everywhere else I used "broad".  Thanks for pointing that out.

[wolf_59]

Wow! Now I'm totally confused, I thought I had a pretty good handle on this from everything I've read on this site and in the TTS manual, This is the way I understood it VE is the measurement of air flow thru the engine at %TP and RPM and that VE will only need a change if you change airflow requirements such as air filter, cams, different exhaust etc.
And once you have the VE's set then you set AFR values where you want them on a OPEN loop system and the ECM will provide fuel needed to meet the AFR values selected for the VE for each cell and on CLOSED loop system instead of changing AFR values from 14.6 you would adjust the CLB's to acheive the AFR values that you want and the advantages of CLOSED loop over OPEN loop besides the ability to V-Tune to set your VE's is that it is constantly sampling the AFR and making adjustments for minor changes in fuel quality & requirements to meet the CLB's that you set.

   

[FLTRI]

Let me see if I sum this volume of information for those who want plain and simple explanations:
Do not use the CLB to change resultant AFR from your closed loop system after the system has been tuned. By doing this you will also change the resultant open loop portion of the mapping. Set the CLB to what you want tuned to in closed loop.
How's that?
Bob

[Don D]

Read Glens right up again carefully. This isn't a topic that gets stuffed into a 2x2 box and on the shelf. Complex but following the TTS instructions to adjust the MAP by VTUNE and the TS for the higher throttle settings works reasonably well. Not the perfect tune by any means though because timing is not even touched yet and the values are just reasonably close.
Don't for get the items that get toggled off and the AFV. These are important.

[glens]

Thank you.

For the sake of this part of what I intend to say, we're going to establish that every external and internal operational parameter/condition will remain consistent so as to eliminate variables and isolate what I'm talking about.  Engine and air temps, relative humidity, the fuel itself are perfectly consistent throughout.  Everything else is unchanging except for EFI programming values.  The programming, especially with the closed-loop functions, will accommodate all these other things as they swing around but I want to isolate some of those functions as the topic of discussion.  So please understand it that way.

This one here left me feeling a little understated when I looked at it again earlier:

So in the end the results wind up not being quite the same; there's a distinction.  If they're achieved with v-tuning, the AFVs will be ~100%.  If they're achieved with "actual" VEs from an exterior source but using the same CLBs as the v-tune would've used the AFVs will be ~103% (or whatever).  AFVs get applied everywhere they can so you need to take this into consideration and tune the outlying areas afterward with as-close-to 100% AFVs as possible.  It's a shame we can't know where the AFV cells are or what they contain.

What I'd meant to convey there was that if the VEs are obtained via the v-tune software with "custom" CLBs, while later running closed-loop the AFVs (one last reminder of my caveats/conditions) will "cruise" right at 100%. 

With CLBs of ~450mV (or whatever value actually provides for "14.6" [pure gasoline] AFR, probably more like 500mV) a v-tune should produce the same (closed-loop-area) VE numbers as an open-loop TwinScanII tuning session.  At least proportionately the same. This will provide "working" AFVs of 100% as well.  When you then later change just your CLBs the AFVs will effectively change the VEs to match, but this effect is rather farther-reaching than might be expected.  It doesn't seem to get talked about very much other than by me.  It's these elusive AFVs and how/what they do which got me interested in all this stuff in the first place, so maybe having that knowledge y'all will be better able to understand/forgive me my folly.

Admittedly, we're only talking about a few percent, typically, here and there.  If you want a good-enough tune and then just ride the damned thing, this can all be dismissed.  All I'm getting at here is that the two paths to "the destination" don't get you to exactly the same coordinates, and if it matters to you, you should take the route that gets you there.  Maybe "in the county" is close enough; maybe "in the city"; and maybe you're looking for a specific address.  Over the face of the entire globe they're all probably enough the same.

Minutiae is my specialty and sometimes it can sound like I'm talking about things much larger (as in "blown all out of proportion").  I apologize for that.  But then again, we are talking "technicalities" here...

Don't think it would work very well for "street tuning", but it would probably be beneficial if on a dyno to get the VEs where you want them (the first time) and then go open-loop with a Power Commander in-line and play with the timing "live".  Now this won't be perfect since the PC must delay the timing event by at least one revolution from when it was "sent".  But I'd bet it would save time getting the whole job done.

It's not really understandable how different spark timing could alter the VEs if you consider the VEs to be strictly "how much air is flowing" in a purely mechanical sense, so maybe it's best not to think that way.

I surmise a good tune requires VE, then spark, then VE again (and for the really anal, another pair in order).

[Steve Cole]

I guess what most people need to do is understand we are trying to get the VE by backing into it. As you look at things that effect how we are trying to back into it maybe that will make more sense to some of you. VE is the total airflow entering A cylinder under a fix set of conditions. The truth is running an engine is anything but a fixed set of conditions.

With that said we used the resultant fuel delivery to work backwards to arrive at VE and anything you change in the fuel delivery can and does effect the outcome. Now how much it effects it is a different story but let's just go from this point. So now that you understand that fuel effects the outcome we have to take care of that condition so we tell you too by rerunning V-tune after anytime the fuel system is changed. None of this is 100% perfect and as the fuel quality changes this effects the system the same. Our goal is to make the ECM happy in as many conditions as possible so that it does not have to make any corrections. You need to remember that each time a correct needs to be made it is done after the fact by the ECM. With the ECM in closed loop, no matter what you or I do it still has a mind of it's own so we need to make sure it's (ECM) happy with what we have done.

In theory with nothing changing once the VE is arrived at there would be no need to change them but the difference between theory and real world will jump up and bite you in the ass every time! As the cylinder head heats and cools or the reversion in the intake changes or the exhaust changes the VE is effected. All these little things and many I have not stated can and do cause the VE to be different and anytime you are measuring with an external device that doesn't know about all these little things your answers will be off. This is because we are all backing into a VE value, it is not being measured directly! All external devices are assuming what is going on to arrive at VE and the assumptions in many cases are wrong!

On a dyno or on the street doesn't matter, we want the ECM to be happy with what we are telling it to do. If it's not it will change what we've done to what it wants over time in closed loop. Once your in open loop operation the ECM quits LEARNING to try to correct to what it wants but it still applies what it knew to be correct from before. So let's say you run the engine and the external device says the mixture is something other than what you want, so you make a correction to get what you want. Now the bike runs in closed loop for a time and learns that it wants a 20% correction to be happy. Guess what just happened to your open loop mixture? It just got a 20% correction applied so now it is not where you wanted it! Same thing happens if you tune to an external device and the ECM is already applying a 20% correction. Since the external device has no way to know about any of the things the ECM is doing one needs to be careful when trying to understand what it maybe telling you.

[FLTRI]

Say one want 14.5 AFR in closed loop.  That is 768 CLB.

Per Mastertune O2 conversion 798 = 14.25 NOT 14.5.
Bob

[strokerjlk]

The ONLY thing that makes sense is that a v-tune... which is a kinda hybrid deal, uses the CLB in ITS tuning process.  While a REAL dyno does NOT use anything like that and derives it's numbers from purely the sensor.

you guys are making this way to hard. LOL when you set up your map put the area you want to run in closed loop to 14.6 . then set the other areas to what you want them to be say 13.2-12.5 100  kpa. 4500 rpm 80 kpa 13.5. with the twincan you are not tuning with CLB so when your done and you put your stock sensors back in set the CLB to what you want  ....then forget about it.
the AFV is going to adj. (make corrections) no matter what you use (V-TUNE, DTS) that's what happens when we run in closed loop.
as far as the twin scan tune getting  20% corrections added in open loop after you change the CLB in closed loop I have never seen this. maybe 2-5 % if you check it later. then if you check it again later it could be 2-5% the other way.
again this is what happens when we use closed loop. I have been using the DTS for 3 years and I have checked and rechecked diff tunes to see where they were at after time. yes there will be some ve's that need adjusted 2-5 % but no matter what method you use to tune this will be the case....again as long as we use closed loop. this is the reason so many tuners still refuse to tune a 07 up bike to closed loop. so if the small variances bother you ,me anyone then don't tune in closed loop. I use both depending on build and customers expectations.

tend to think a TTS v-tune THEN a TS for final probably works best.   But... then again maybe not.  If the tools are avail;able, why not try the different configurations and SEE what is arrived at?  MAYBE a bike street tuned with TS will be the better running tune?  WHo knows, right?  On any given bike, especially.

John you paid 500.00 for a piece of equipment to be able to tune your bike without making any guesses. use it for that. I have used both and I don't have to guess what works the best. OR 2nd best to a 4 gas analyzer LOL
after you get it and see what it does you wont have any of these doubts anymore.
I have said this before but it bears repeating......I tuned bikes on the street with a twinscan before I got a dyno. I used it for AFR and timing. then took the bike to my local dealer paid 32.50 for 3 pulls with each map I created. I usually made 3 maps for each bike so 9 pulls total. then handed the customer the sheet when he picked up the bike. Funny thing is.... I was getting better numbers and better drivability then the dealer was on there dyno tunes (stage 1 & stage 2 bikes). this came from one of the owners as we examined sheets.
I have had a few of these bikes back on my dyno since road tuning them. I was able to get a little more out of them (with timing) but not much.

[FLTRI]

Wouldn't V-Tune be an external device, too?  But the MATH is integrated?
[/quote]
I think, since the changes are all based and made from internal ECM data, this can be called an internal tuning system?
Bob

[FLTRI]

you guys are making this way to hard. LOL

Over-thinking seems to be the norm for some. LOL!
I also rely on the O2 readings from the "pipe" after tuning in closed loop (TTS) to verify the actual AFR and guess what? They are always within a .1 or .2 AFR at the most. Having the ability to check/recheck my tuning with the dyno AFR sampling is more confidence inspiring than "hoping" the system performs as advertised.
Bob

[strokerjlk]

you guys are making this way to hard. LOL

Over-thinking seems to be the norm for some. LOL!
I also rely on the O2 readings from the "pipe" after tuning in closed loop (TTS) to verify the actual AFR and guess what? They are always within a .1 or .2 AFR at the most. Having the ability to check/recheck my tuning with the dyno AFR sampling is more confidence inspiring than "hoping" the system performs as advertised.
Bob

exactly
and that's the best we can do with what we have to work with....whatever that might be
could get into the great sniffer accuracy debate here if not careful         

[FLTRI]

...could get into the great sniffer accuracy debate here if not careful         

I'm ready! LOL! I've been verifying O2 readings from the bungs against the "sniffer" readings for over 10 years. It started from posts stating one is more accurate than the other. I have found instances where the sniffer produces more reliable readings than bung mounted O2's, usually due to exhaust system choices that the bungs were not in ideal spots/angles which produced erroneous readings from reversion.

Since the sniffer can be relocated in the pipe the reversion sometimes can be eliminated from readings. This luxury is not available if the only method of reading come from permanent bungs in the pipes.

My experience tells me to check one against the other, especially if there are tuning issues (ie: low speed, light throttle running)
Just my $.02,
Bob

[strokerjlk]

My experience tells me to check one against the other, especially if there are tuning issues (ie: low speed, light throttle running)

or center fold pipes.
got a lot of time and money invested in homemade gadgetry.....to get accurate idle readings on baffle-less pipes.
just saying.....the .1-.2 differences  we were talking earlier could now be .2 -.4  when a sniffer is used .still not enough to panic.
same could be said if we are re checking with sensors in the head pipes if the original tune was done with a sniffer. to many variables to see them ALWAYS exact even if both are installed at the same time. been there done that also! LOL.... all's good no debate

[wolf_59]

Let me see if I sum this volume of information for those who want plain and simple explanations:
Do not use the CLB to change resultant AFR from your closed loop system after the system has been tuned. By doing this you will also change the resultant open loop portion of the mapping. Set the CLB to what you want tuned to in closed loop.
How's that?
Bob

That even I can understand
Thanks, Bob

[Steve Cole]

Wouldn't V-Tune be an external device, too?  But the MATH is integrated?

Can one SEE what the AFVs that have been applied?  I still have v100 of Matertune (been a customer awhile, I guess)... can THAT tell me what AFVS have been applied?  Curious...  Maybe use the TS and see what the bike is outputting at any given moment would give one a guesstimate.?

Since we do everything from the decisions the ECM makes we are not an external device. The ECM makes it's mind up from all the inputs, so it knows everything that is going on. We just take it's answer and adjust it to where we want it to go. In this way we make sure the ECM is happy first and foremost. An external device can only see the final output from the ECM with no idea of how it got there, it cannot tell the difference between Decel, Cruise or Power Enrichment modes, it cannot know what the BLM, Integrator or AFV is at, so it's up to the operator to try and see what mode the ECM is operating in during the external device tuning.

A good dyno tuner will have learned there is something different going on in some conditions and disregard what the external device is telling him, but to an untrained/unexperienced person they do not have a clue as to what is happening. I have seen this time and time again when going out in the field and watching what some dyno guys are doing. Let's just say that there are a lot of dyno operators and very few good tuners out there. The training HD dealer tech's have received from HD is very very poor to say the least.

You cannot view or see the AFV in any version of DataMaster let alone know when the data is good versus bad. There is just too much going on within the ECM. That was the reason for V-tune, it can compute all the data better and quicker with many less errors than I can myself and I've only been at this for 30+ years.

[FLTRI]

Bob, do you have some problem with me?  768 IS 14.5....

OOOOOOOOOPs!! Soooory, just another dyslectic maneuver on my part.
Bob

[strokerjlk]

use the TS to see about 'blending'

.
there is no blending with  twin scan. you hit all the cells and make corrections. same way it was done before v-tune came about.
think about it we were tuning closed loop with SERT without V-Tune.
set the decel, accel to 0. disable the pe and collect data.
we had no way to disable pe with sert so we had to start leaner so as the pe didnt go rich to soon and kill power. that was just a way around something we can now control with the TTS. so many things going on we don't have control of.but we learn to deal with or get around it.

[Don D]

Listen to the man!
Good advice.

[Steve Cole]

So... taking Steve's statement out to their logical conclusions... Folks with a TTS MAster Tune have NO need for a dyno operator.  If one can take the time to hit ALL the cells properly, turn them green, get the check mark... what good is a dyno?  I can personally do one thing a LOT of folks can NOT.  I can get to some SERIOUS hills fairly quickly.  With a monitor, and hills, brakes and throttle... should be able to get them ALL green with the 80kpa check box inside.

This is a SERIOUS question.  Steve is passionate about this, and states a good case... it ALL needs to be done internally with the ECM controlling the 'play', what's left?  The 'tricks' that tuning experience can bring, I would assume, ie; reversion?    Also, using a TS for top and bottom kpa... what else?  I've read his statements a few times... and the case is: NO dyno operator needed.  Hmmmmmm... don't know if I go there, personally.  But... what I do for a living is program PLCs... i 'get' more than one would assume from my posts.  Steve's comments make me think of the TTS Mastertune's 'selling gimmick'.  That an OWNER, withOUT a dyno can get 90% as good a tune.   Listening to Steve, makes me think that last 10% is where blending occurs, right?  A 'dead on' TTS tune just HAS to be better than the dyno operators that Steve talks about.  Doubt it's better than Tuners like Jim and Bob, though.....

A good tuner that has learned all the problems will be able to tune outside of closed loop better than V-tune will but not within the closed loop area

I just MIGHT fall into the Don Dorfman camp on this.  Ultimately... a 'hybrid' tune is going to cover the most bases in the best manner.  No 2x2 box here.  Maybe an Urn, though.

What I do know, with CERTAINTY is this:  my bike (with the old top end) would CRUISE right at 30kpa.  Now maybe it REALLY wasn't, but that's what the MAP sensor said, and the ECM goes off of that, and NOT reality.  Steve has said that below this KPA... v-tune doesn't function.  Nor.. does it function above 80kpa.  I have purchased new plug wires and a coil.  I will run the proper heat plugs.  TRY to make the best TTS environment as possible.....


I have never said this, so please quote me correctly. The lower limit of learning in the ECM is 26 kPa and the upper limit is 83 kPa. This is of coarse when using Mastertune as it is not the same in all HD calibrations.

I'm on THIS site for one reason... get a good tune for my bike.  I argue, etc, as MY way to learn things.  From reading Glens and Steve... makes me wonder where MY tune really is at....  Jim states a good case, too, AND it seems that Jim is one of a few who wish to tune a DBW closed loop bike.

MY first attempts to fix my tune will be to use the TTS to its fullest, and then use the TS to 'check and verify (like its the Soviet Union) those settings and then use the TS to see about 'blending'.  Make THAT map the absolute best it could be, I guess.  Come on here and ask more, obviously stupid, questions and see what the true result is.

THEN... try the TS alone.

Compare.  Report.  Learn.

Steve...I started a new thread for you to post in.

Dyno tuning done correctly is great but you first have to know what your doing and then what the ECM is doing. That knowledge is not something that is in any book that I know of and it's not always the same for each motor build. It takes time to learn these things on many different configurations. V-tune will get you about 85% of it knocked out very quickly but you must do the rest of it with the tools provided with Mastertune. You must watch what is happening, learn, adjust and move on.

[glens]

you guys are making this way to hard. LOL ...

... after you change the CLB in closed loop I have never seen this [20% (which was not a suggestion of what might happen with what [at least] I've been talking about -- that 20% was merely a figure thrown "out there" "for example" by Steve)]. maybe 2-5 % if you check it later. then if you check it again later it could be 2-5% the other way.

It's not too hard for me.  Hahaha!  I'm sure I'm not so tedious-seeming in everyday life, but this is the way I approach just about everything.

You no doubt saw where I was throwing figures like 3% around.  Heck, that'll show up on subsequent v-tunes for crying out loud.  That's why I (ultimately) stipulated exclusion of typical operational variances in order to isolate what I was talking about.  (What'll happen to the "upper" open-loop areas after tuning them and then changing the CLBs.)

You maybe also read me say something about this possibly being equivalent to varying, very tiny, degrees of pinpointing something on the surface of the earth.

Ultimately I'd hope folks would come away with the ideas as I'd intended to express them and tried (probably largely unsuccessfully) to iterate them.

Tuning with the SERT had to be problematic enough without the "factory" version of the software prior to the advent of our closed-loop.  TTS has since made available an excellent tool for getting the VE information very largely correct.  It really is folly (at least the way I see it) to not use it as the primary tool.  Steve and Bob have more eloquent delivery than I do but the way I read what they're saying, well, it's exactly what I've been saying.  (Do I say that at the risk of offending them?)  There are several possible ways of saying the same things.  I like to sum it up by the notion of "getting the VE tables so the working AFVs are as close to 100% everywhere as possible".  Or characterizing v-tuning as "centering the AFVs under normal operation".

I'm a big fan of our Delphi system.  I like the way it works.  I like to have the most adaptive "headroom" available in both directions.  I feel I need it when I travel.  Having it well-centered is akin to having a full magazine of fresh ammo in a gun.  Having one in the pipe, too, would be nice, and for private street tuning, following up a comprehensive v-tune with a TwinScanII+ (or whatever it's called at this time) for the areas left unchecked is about as good as it can get for arriving at good initial VE values on the street.

So long as all the trouble is being taken to do this privately, it just makes the best sense to do things in the proper order.  It should be quicker and more accurate.  Why intentionally dick things up 3 or 4 percent in places by reversing the order of your activities?

I've said this many times in many places (probably would be really embarrassed if I ever found out just how much time was spent doing it!):  Get your system so the AFVs are well-centered ("utopianly" 100%) in service.  This way, you'll be best "ready for anything" as you can be while you're out using the bike.

Damn!  I just can't "bale hay" worth a damn!  Bob?  How do you do that?! 

[Steve Cole]

If that does not work out remember Velcro is your best friend.

[glens]

Yeah, you can see that screen in the garage!  Might not be worth much in broad daylight?