Ok the results are in.
Firstly, this is the curve I tested. As noted above, it is the result of breaking down 16 runs into 1K sections and determining the best advance values for each 1K section.




Snap to WOT from 4K in 3rd gear. Measurement of time between 5K and 10K in seconds - 120fps 720P video. I list the above curve as "BD" below.

Curve Run1 Run2
-----------------------
25° 19.95 20.07
BD 19.53 19.53

 

So this "break down" (BD) curve is quicker. When viewed @ 120FPS, it is about .5 seconds quicker - I don't know what this would be in "real time". I can say that the bike is more responsive, ergo powerful, everywhere - I'm guessing something relatively small in extra hp everywhere. Perhaps 1/2 - 1 hp more everywhere or in most of the RPM range.

I did test a third curve tonite that did not produce good results. I will break down all three curves into small pieces to further analyze the performance.

Inch by inch, I'm getting closer..

 

OK so I have been doing some tweaking and testing and have got to the point where I think a dyno would be required to optimize further.

This is the curve I ended up with, very similar to the last curve you posted.


feels like I got a decent boost in the Low and High ends and not too much extra in the mid range. Off road with the tall SF gearing, the extra grunt is a huge help. Much less slipping of the clutch was required on slow rocky climbs, hopping logs etc. Also pulls much harder in the higher rpm 9-12k range.
Seems like the stock advance was pretty close in the engines peak efficiency range so I doubt max power has been impacted much.

 

Yes, the large increase and then "slope down" on the "front end" of the curve provides a big jump in throttle response and power ramping below 6K - I will get around to finding the best shape for that "end" of the ignition curve.


For now, I'm going to continue to work on finding best values for the "peak efficiency range" (6k-8k) as I think this is where a significant power increase can be found. Also, it's easier on my bike to only rev 4k-8.5k @ WOT for such testing..

Then I'll work on the other two ends of the curve.

When I achieve a 1 second increase 5k-10k, (using 120fps video) I'm gonna dyno to get the data and determine a "next move" for further testing..

 

Yep, it's been awhile since I posted here.. Well it's really hot now in N. Texas (100°F) and I don't like riding during our summers.. BUT, I have been "at it" , and have "data reduced" another 20 or so runs, compiled a new curve based on all findings, and tested it..

The data, on these hot evenings, is not producing blatantly obvious choices at a lot of given RPM points like it was during the cooler months' testing. I am also having a weird situation whereby the stock 25° curve will "out of the blue" create numbers on a run that mimic the more powerful curves and advances I am testing (See run 2 below). I take great pains to create continuity between each run - it almost seems like the module decides to run my advanced curve instead of the stock curve every now and then.. I have had to run this comparison test twice due to this problem, and still I get an "errant super performance" 25° curve result while testing today.. Oh well.. here's the results..

This is the Curve created from all test runs (about 40). I now have data 5-10K @WOT 3rd gear, that spans 25° - 30.5° in .5° increments.

I tested the below curve against the stock 25°.


And here is the results from analyzing the 120FPS videos of the instrument panel, measured in increments of .03 seconds.

RPM Interval______Curve___Run#__Time

5K-10K---------------25°------R1 20.25--R2 19.59(WTF?)- R3 19.77
----------------------TEST------R4 19.56--R5 19.56-------R6 19.38

Breakdown:

5k-8k---------------- 25°------11.01----10.47---10.83
-----------------------TEST-----10.53----10.53---10.56

8K-10K----------------25°------9.24-----9.12----8.94
-----------------------TEST------9.03-----9.03----8.82

I have no idea why the 25° curve produced such varied results - my curve produces quite similar values.. Nice that there is almost a 1 second difference between the best and worst run between the two curves..
This was 6 runs performed back to back without stopping the bike - 3 runs with 25° followed by 3 using my curve. Wind was calm and I sat in the exact same position for each run using the the exact same course. I am at a loss to understand the better times on run 3 (25°) and run 6 (my curve) if you throw out that weird run 2 result..
Comments/Ideas welcome.

 

It is interesting to see that your results were very consistent.

Would it be worth it to test it against the stock cdi module rather than the stock curve in the programmable module?

Do you have a general sense of weather the this mod is worth the money?

Is there a sense that every setup is going to require a custom map. I am running a TM36/351/Q4. Do you think your map is going to be transferable?

 

Tax, most of the time I get identical - or statically identical - numbers.

Wide variances between, say 3 runs of 25°, give me real headaches. I have no idea what to do with that data - except get mad about the time and effort wasted. I will continue working on this "variance" problem until I have it solved.

Is it worth it? Well that's what I'm trying to figure out for all of us. The increase in low RPM power ramping and throttle response (2k-5k) with the "big front end" on my test curves is really nice/possibly worth it - but as of now I've only measured it with my Dyno Butt and roost marks on a hillside. I will eventually concentrate on this RPM range to get better numbers/shapes..

My curves will work on KLX's modded like mine - Yours could use/benefit with the "big front end" but would certainly require something different from 5K on up.. If you get the equipment, I can get you started.

 

KLXster, best thing you can do is take a couple of curves you feel most confident in and dyno them compared to stock timing. Youve probably hit a wall as far as repeatability with what youre doing.

Do you ever get the high rpm misfire/ single spark cut when pushing your bike hard through the gears? Did you ever?

 

Next dyno session is when I get a 1+ second increase over stock 4-10k.


I do get single spark miss @ high RPM's from time to time - haven't had any during testing tho...

 

Might not be possible to get there though, with just ignition advance adjustments alone.

Wonder what the heck causes those single spark misses....