Print

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

[mayor]

Ok so we have it established that a steady voltage can be read at WOT, how do we DIY tuners determine what the voltage in that instance transpires to in AFR without some other sensor to tell us? 

great question.  The posted data suggests that using the data collected from narrow bands while at wide open can be used to move the ve's in a manner to produce a somewhat linear sampled afr line, but I don't think there has been a way to tie that into the original stated purpose which was to give the DIY tuner the ability to target a safe WOT range.  based on that, let's move try to get to that point:


Bob,

since you are reluctant to post your afr data, due to fear of muddying the waters.  Can you at least post the proposed target mV values for each model calibration that you have tested.  At least with this, those of us that are willing to post data can see if there is a correlation if we get a chance to test those models.   

[whittlebeast]

I'm assuming that this is directed at me, since Joe is a relative new comer to this conversation, so let me respond by asking in what way have I been fighting this? 

You seem to have a rather short memory.  When I made the jump into flash tuning a few years ago, did I not go through the process as outlined by TTS and its representatives, all the while documenting publically what I was doing, with a final outcome that validated that vtuning per the outlined process worked?  In the validation, did I not also publically document that when the Vtune program on an afr bike indicates that enough data has been collected on an extendable cell that the results are fairly linear to what was recorded on an external afr meter?  The lambda cals are the only one's that I do not have confidence in the extension method, and I have not seen any circumstance to get me to change my opinion on that. 

I am not opposed to using broad bands, but I am not opposed to using narrow bands either. In the several attempts that Bob had made previous to get this thread going, I never publically stated that it wouldn't work. I honestly didn’t know, in truth I was actually biased towards the notion that it would work since Bob and Jason both had stated that WOT tuning using narrow bands was not a new concept.  If I mind correctly, it was you telling Andy Whittle that narrow bands can't be used in this fashion several years ago, not me. 

The only issue that I have on using the narrow bands to tune with is an operator at this time is not dealing with known values.  If you go tell someone to target 1720 mV on a lambda bike, they will look at you like you have two heads.  There is nothing to tell them how many ve values that you must increase the cell to if the current reading is 1610 mV.  The operator can use afr since the calibration has a table with desired afr, and then the math between desired.

WOW, I can't believe I got drug into this after all these years.  The tuning I do now has come a long way since those days.  Most of the motors I mess with now days are making about 2.5 - 3 HP per CI.

Have fun tuning

Andy

[mayor]

hey Andy, welcome to the conversation.

[whittlebeast]

Here is something I put together about a year ago.  Much of this was was discovered when I was messing with the Harley.  It has applied to every motor and tuning system I have messed with since those days.

http://www.ncs-stl.com/tuning/

Have fun tuning

Andy

[FLTRI]

Andy,
Good hearing from you. So looking at the scatter graphs I see what you are talking about. Tuning issues all over the place.
What did you do in each case to correct these obvious running issues?
Bob

[whittlebeast]

So looking at the scatter graphs I see what you are talking about. Tuning issues all over the place.
What did you do in each case to correct these obvious running issues?
Bob

The first thing I do now is plot Duty Cycle vs MAPxRPM/100 and look if the result is a reasonably straight line.  On the other plot I do MAP vs RPM.  I can then I tag the points that are out of the trend line to see where in the data in the MAP vs RPM plot corresponds.  The trace will also automatically sync with the two scatter plots so you can see if there is any pattern to when the data does not appear to be behaving.  Way cool stuff.

The other new thing in the SW does now is data filtering.  You can easily throw out selective data.  For example, only give me data when the bike is in 4th, 5th or 6th gear.  Or for example throw out the data when the throttle is above 5% and the motor is cold.  The options never seem to stop.

Once the filters are set up for any given system, I can easily find the problem areas in a tune in a mater of a few min.  The link I posted above gets into lots of this stuff.

Andy

[FLTRI]

Once the filters are set up for any given system, I can easily find the problem areas in a tune in a mater of a few min.  The link I posted above gets into lots of this stuff.

Andy

So are you correcting VE tables? What else? ...for these problem areas?
Bob

[whittlebeast]

Generally, yes the VE tables.  You may find things like extreme reversion making the O2 to get the wrong answer.  Bad MAP read angle in the code....

Who knows until you look at the data.  I have run into this several times when people mess with the intake length.  It's been several years since I even looked at Harley data.  Most of this new found data analysis capability has come up after the SW was changed.

I got sucked into helping  do a Harley turbo Megasquirt install right now.  I will know lots more about Harleys in the next few months.  The data logging of the MAP can get down to the tooth read resolution if we need it.

Andy

[Steve Cole]

Here is a dyno result and AFR from tuning using only the Narrow Band factory sensors.

[whittlebeast]

How does the AFR overlay when you do say 30 pulls in various gears like happens in the real world?

Andy

[Steve Cole]

Does just fine, NO different than when tuning with Broad Band sensors. The repeatability is the same either way. The truth is nobody does 30 WOT pulls in various gears in the real world. At best for most riders they are lucky to do 3 -4 WOT pulls as they run it up through the gears! Very few, if any, run a HD through anything more than 1 - 4th gear at WOT, at any one time. Those that are road racing are typically using only 3 gears on the track and there isn't hardly any HD road racers to speak of.

So in the real world of a HD engine your lucky to get someone who runs them WOT from 1st through 4th gears!

[whittlebeast]

I used to get all sorts of data logs off Harley installs that had a huge cam.  The log of a 60 min ride never got over 2500 RPM.  I never quite got my head around the mind set.  CTS1950 used to rev/beat his motor to no end.  He was fun to work with.

Engine RPM sweep speed on narrow bands and the resulting exhaust temp was why I went to watching the PW and DC at high power situations.  I just used the NB sensors to Identify where the motor was having trouble holding consistent AFR.

Have fun tuning

Andy

[whittlebeast]

Here is a little data log off of a 1500 cc, 260 HP, 8100 RPM blower motor I have been messing with lately.  On the right is the MAPxRPM/100 vs DutyCycle.  On the left is the typical MAP vs RPM vs AFR plot.  This is all done with a PC5/Autotune/LCD-200 setup.  The mess at the top right corner is where the injectors are going non linear at 92% duty cycle.



http://ncs-stl.com/CMD/BasicTuningView.png

Andy

[Steve Cole]

This is all done with a PC5/Autotune/LCD-200 setup.  The mess at the top right corner is where the injectors are going non linear at 92% duty cycle.

Andy

There is a very good reason you are not suppose to run injectors over 80% duty cycle and you are just showing what you not supposed to do. There are limits and when you go past them, things are not what you would want them to be.

[whittlebeast]

Yep Steve, it is hard to believe the factory set it up that way...  Who would have thunk it?  That motor had a PC5 for it sitting on my desk before the I picked it up from the dealer.

Issues jump right out at you when you have the tools to spot them.... 

Andy