The sort of thing that gets my attention

[Jeffd]

how would the bt in post 25 look in3d? 

[whittlebeast]

how would the bt in post 25 look in3d?

[Jeffd]

better  yet what does it all mean/show?  thanks jeff

[PC_Hater]

better  yet what does it all mean/show?  thanks jeff

Those scattered graphs mean that the bike is not in an optimum state of tune.
Compare the smearing in the graph above your post with the relatively clean dots in the post above that.

As to how you improve the 'smeared' bike - dunno. Fit a carb and forget all about it seems best...

[Doghouse]

I create attractive but meaningless graphs for my company every day.  Unless WB starts explaining them, they will remain just pretty pictures to most.

[Jeffd]

better  yet what does it all mean/show?  thanks jeff

Those scattered graphs mean that the bike is not in an optimum state of tune.
Compare the smearing in the graph above your post with the relatively clean dots in the post above that.

As to how you improve the 'smeared' bike - dunno. Fit a carb and forget all about it seems best...

So post 49 looks cleaner then post 51? just trying to understand this stuff lol.

[whittlebeast]

I create attractive but meaningless graphs for my company every day.  Unless WB starts explaining them, they will remain just pretty pictures to most.

Engineering is all about finding what things are functions of something else.  And recognizing/determining what is the cause and what is the effect.

This is the sort of thing I am looking at in a system that has high resolution on board data logging.  This is at 100 data points per sec.  The spikes you see in the MAP is driven by misfires in the ignition system showing up in the logs.

http://www.ncs-stl.com/mustangms3/SP32-20110607-055216.jpg

AW

[whittlebeast]

This is cool.  Look how the AFR gradually fades from 14.6 to 12.5 as the power comes on.

[Sporty 48]

The 74 cubic inches of the Sportster are inefficiently used even for an old wobble fire design.
Doing other Sportster things but paying attention.
About time to wake up the Sportster.

[FLTRI]

Looks like there the engine runs richer in lighter loads than heavier loads?
ie: ~27/40MAP @ ~2pw
Still looking for your explaination of how these scatter graphs make it easier to tune...as a tool.
Bob

[whittlebeast]

I guess I am a little confused a what is throwing you.

PW  is essentially torque.  Duty cycle is essentially HP.  The higher the PW or Duty, the higher the target AFR should be.

Timing is generally driven by the speed of the flame front and the speed of he motor and to some degree the air fuel mixture.

Detonation is generally the spontaneous combustion of hot fuel and air under high heat and pressure.  It must be avoided at all cost.  Locating it is the first step in avoiding it.

Data logging is simply the tool that we use to look for the patterns.....

AW

[whittlebeast]

Check out the difference in these three timing maps.

Stock Sportster timing map


Stock BT timing map


My Sporty Timing map

[cts1950]

I have been playing around with the scatter graphs. I have been looking at fast retard events and this shows them clearly. the cursor on the left graph is showing 3068 rpms and 65.8 kpa and the red tone of the dots show a 8 degree fast retard. This helps you to zoom in on a entire data run and see the problem spots.

[attachment removed after 60 days by system]

[PC_Hater]

better  yet what does it all mean/show?  thanks jeff

Those scattered graphs mean that the bike is not in an optimum state of tune.
Compare the smearing in the graph above your post with the relatively clean dots in the post above that.

So post 49 looks cleaner then post 51? just trying to understand this stuff lol.

I think that post 49 looks cleaner than post 51.
There is an immense amount of information in this thread. It took me more than an hour to work out what was going on so I could make a 'sensible' first post and not make myself look a complete idiot. And then the later graphs and comments add yet more info! Happily my EFI bike is MM and my other bikes have carbs so I don't have to worry about this stuff too much. My girlfriends EFI Sportster runs a bit too rough at cruise for our liking so I'm reading these threads as a primer in case we have to get the thing tuned just to make it run right in standard factory trim.

I suppose I should add that with a background in electronics and software I know that if I put the effort in I will understand it. If you don't have enough science or maths or practical hands on experience then it will be a very steep learning curve. 

[Sporty 48]

Whittlebeast,
By the looks of your tuned Sportster and the stock BT, the Sportster has come a long way and still has a bit of improvement to go.
Is the Sportster near as good as it can be?
What % of the table is the Sportster closed loop?

[FLTRI]

I guess I am a little confused a what is throwing you.

PW  is essentially torque.  Duty cycle is essentially HP.  The higher the PW or Duty, the higher the target AFR should be.

Timing is generally driven by the speed of the flame front and the speed of he motor and to some degree the air fuel mixture.

Detonation is generally the spontaneous combustion of hot fuel and air under high heat and pressure.  It must be avoided at all cost.  Locating it is the first step in avoiding it.

Data logging is simply the tool that we use to look for the patterns.....

AW

This opens an entire new world of easy data analysis and tuning EFI.

I'm thinking the old DataMaster recording for timing works very well at seeing detonation issues.
Just not seeing where your above statement applies.
Bob

[whittlebeast]

I have been playing around with the scatter graphs. I have been looking at fast retard events and this shows them clearly. the cursor on the left graph is showing 3068 rpms and 65.8 kpa and the red tone of the dots show a 8 degree fast retard. This helps you to zoom in on a entire data run and see the problem spots.

CTS1850  Here is how I like to look at a bike with knock issues.  Notice when your bike has knock, the timing is already pulled back.  This is really starting to look like a serious mechanical or cooling issue.



Note that this all looks far better on the screen as apposed to a screen "Potty mouth" that looses all sorts of detail.

Hope this helps

AW

[whittlebeast]

FLTRI

I realize that some people are skeptical that we can learn anything from scatter graphs.  I have been messing with these for the last couple of days looking at data logs the I received from quite a few different people.  I find the data logs from the stock bikes and their timing maps the most interesting.  At times, the factory maps defy logic.  There must be underlying logic but I have yet to get my head around it.

On paper, the Sporty is very similar to the BT motors.  I would have never expected to see so much difference in the base maps.  As in all of engineering, it tends to get messy at first but eventually it get easier once you realize what things are functions of some else.  Often the relationships are difficult to find and then document.

Have fun tuning.

AW

[FLTRI]

FLTRI

I realize that some people are skeptical that we can learn anything from scatter graphs....

Not skeptical just trying to figure how the time needed to properly tune a bike can be sped up by studying these graphs.
Been asking the same question but no answer. I agree these graphs are very interesting and can provide quite a bit of understanding what is happening, but that was not my concern. It's all about always being open to new, more efficient methods of tuning. I have worked very hard over the years to develop tuning procedures and techniques that net better tunes in less time.

I find the data logs from the stock bikes and their timing maps the most interesting.  At times, the factory maps defy logic.  There must be underlying logic but I have yet to get my head around it.

You must understand stock calibrations are developed to pass emissions regs NOT best performance.

After some 12+ years of observing and learning how these calibrations work, not to mention pestering SC for many years it has become very appearant there are strategies strictly developed for reduced emissions ONLY in the stock cals...some of which disregard logic and performance strategies.

...once you realize what things are functions of some else

This is preciselywhat SC has been preaching all along...usually when someone notices and posts observations that the result was not as expected.

Bob

[whittlebeast]

Yep.  This have everything to do with the relationship between ALL of the tables.  Most of us are in fact tuning for the best overall package regarding creating a good running total package.  Decent emissions is a byproduct of a good performance/overall tune.  This method has the ability to find places when tables are fighting each other and causing non smooth flowing plots.

AW

[FLTRI]

...This method has the ability to find places when tables are fighting each other and causing non smooth flowing plots.

AW

Ok then. I think I got it now. You weren't meaning actual tuning becomes easier...what you meant was when you can't get a bike to respond as you expect, you can use these scatter graphs to figure/see out why?

Next question: Once you have discovered "tables are fighting each other" how do you change/correct it? AFAIK, unless you are privy to the code and can actually change it you are really looking for a way around the perceived issue(s)?

Bob

[whittlebeast]

Sometimes it is easy as when the closed throttle timing map is very different from the main timing map.  Other times it is in the Harley Hidden Tables and without access to the code or the hidden table, the only choice is "try a different base map".  This may help to idnetify the issue (or at least find it) that is the root of the problem.  Just think about the the tables that were uncovered in the MT8 versions that in the days of the SERTs were kept a secret.  Things like the power enrichment logic that was brought up about 3 years ago.  In the SERT days, it was a Harley hidden table.  This stuff would make that sort of thing jump out at a trained eye.

AW

[FLTRI]

Sometimes it is easy as when the closed throttle timing map is very different from the main timing map.  Other times it is in the Harley Hidden Tables and without access to the code or the hidden table, the only choice is "try a different base map".  This may help to idnetify the issue (or at least find it) that is the root of the problem.  Just think about the the tables that were uncovered in the MT8 versions that in the days of the SERTs were kept a secret.  Things like the power enrichment logic that was brought up about 3 years ago.  In the SERT days, it was a Harley hidden table.  This stuff would make that sort of thing jump out at a trained eye.

AW

Still looking for these scatter graphs to be the holy grail for tuning but am getting the feeling you may have meant something more like these scatter graphs can be used for diagnosis and studying why things happen.
I was looking for the "much easier to tune" portion of your statement.
My misunderstanding/misinterpretation.
Happy scatter graphing,
Bob

[whittlebeast]

Here is an example of a typical motor with a fairly peaky cam/exhaust.  The motor has very different fueling requirements at higher RPMs.  It turns out that this motor is running closed loop till about 55 KPA and close to stoich up to about 80 KPA.



Based on the MAP vs RPM the rider is running the motor quite hard.

AW

[Steve Cole]

Andy

While scatter graphs look all nice and pretty, please explain how this is going to improve tuning or improve the tuning process to help the DIY guy let alone a professional tuner get there job done?