ESE's works engine tuner

[Wos]

"

Completely closing throttle earlier than normal, however carb is not so tiny at 30 mm and no issue on all gear runs on a dyno which fits very well to our acceleration times on track. "

My guess for koenichs problem

Idle jet to big/ rich

and/or

cutaway to small/ rich

This may dropp pipe temperatures maybe so deep, that pipe is out of resonance !?

Next depending issue could be, that rpm s are dropping to such a low level, where ignition advance is high and this is resultig in another lower egt ...trapped in double issues!???

Wobbly ? He uses a pwk type carb... without tps...only 2d ignition curve

are these other possible reasons for phenomenon??

Thanks for advice, opinion and help !

[wobbly]

The Mychron 5 needs an Expansion Box to run suspension position sensors , and according to its manual this has a set sample rate of 10Hz.
How do you get 25Hz, but apparently even this will not detect chatter as its way too fast a rate of change.
Aim has a software update that allows a Fourier Transform into the first derivative of velocity, but according to their own information we need up to 500Hz to see
high speed rate of change correctly.
https://www.datamc.org/data-acquisit...city-analysis/

Re the jetting issue - it was not obvious at all to me that following a slower bike you would " back all the way off ", but anyway, surely the idle and transition jetting has been tuned
such that there is no issues at idle or on initial acceleration.
Im guessing that you have tuned the main, on the dyno, or on the track to suit the bowl fuel level height after a period of WOT.
In the circumstance you describe, maybe the bowl level is dropping at WOT, and when you back off, as no fuel is being used - the bowl refills back to its nominal height and then when
cracking open the gas again the fuel curve is way rich.
This means the float valve is too small - Easy to check by adding a small breather fitting into the bowl nut and running a tube up the side of the carb.

I had that exact issue using Lectron HV40 carbs with clear bowls on a TZ400. On the dyno when hitting WOT you could see the level drop all of 20mm by the end of a dyno run.
The carbs had 3mm valves, fitting the so called high flow 3.5mm versions with side squirt holes, the issue went away instantly.

[F5 Dave]

An age ago on a mikuki i tried bigger valve etc to try solve a problem where upright you could just make it to the end of straight at Ruapuna B but would die as soon as you opened the throttle after the corner.
I ended up soldering an extension on the bottom of the button style jet in frustration. That fixed the symptoms kinda.
But the real reason turned out to be that the cross carb drilling was the actual bottleneck.

Once that was drilled out (and plug the brass ball again) normal jets could be run.

And that wasn't the last time I encountered that.

[koenich]

True - it's only 10 Hz and yes you need the expansion to add analog inputs, for the basic setup (and if you're not really good with interpreting the data) it's plenty capable in my opinion and was my "budget" solution compared to the likes of a MXM or others with high resolution. To get the velocities I added a math channel with the first derivate of position and it easily went up to 400 mm/s - which either way is too much acc to the information I could gather.

Guessed right - all circuits have been tuned, main been done on the dyno. it rides perfectly rolling on the throttle out of corners, main seems to be spot on as long as I'm on my pace. As its a 50cc obviously WOT is used quite a bit so your theory makes sense and thats also what I feel - that it's uber rich if I'm not pinning it till braking point but rather coasting. I'll check that - thanks

[wobbly]

As I understand it now, even if we see a full stroke dive under brakes at 400mm/s in 1/10th of a second, then there will be only two data points, the start position velocity, and the end position.
With chatter there could easily be 4 or more cycles, up and down in that time frame, with no intermediate data points to show this.
Thus only one peak velocity point will be displayed, instead of what is actually happening - the front wheel going spastically up and down at a high frequency above the sample rate.
In the digital world the Nyquist–Shannon sampling theorem, states that to accurately capture a signal, the sample rate must be more than twice the highest frequency present, thus I can absolutely believe
that to display high frequency cyclic position changes that 200Hz could be regarded as a minimum.
Doing the sampling and data logging is easy with a cheap stand alone processor board - but then integrating that into the Mychron with a CAN line is beyond me.

[koenich]

Well explained, found the same answer while trying to understand the limitations of my setup in the last days.

Besides the MXm there is also the option to use a Solo 2 DL with an AiM Analog to Can Converter which has a sampling rate of 200 Hz.

[Storbeck]

As I understand it now, even if we see a full stroke dive under brakes at 400mm/s in 1/10th of a second, then there will be only two data points, the start position velocity, and the end position.
With chatter there could easily be 4 or more cycles, up and down in that time frame, with no intermediate data points to show this.
Thus only one peak velocity point will be displayed, instead of what is actually happening - the front wheel going spastically up and down at a high frequency above the sample rate.
In the digital world the Nyquist–Shannon sampling theorem, states that to accurately capture a signal, the sample rate must be more than twice the highest frequency present, thus I can absolutely believe
that to display high frequency cyclic position changes that 200Hz could be regarded as a minimum.
Doing the sampling and data logging is easy with a cheap stand alone processor board - but then integrating that into the Mychron with a CAN line is beyond me.

Apologies if everyone is already well aware of it but rather than starting with a dash/display and adding an expansion, one can start with an evo4s and add a display and have much more capable data logging, 5 channels at 1000hz plus you can still use the expansion through CAN to get more if needed , for sort of around the same cost (a bit more but not huge increase), and still be in the aim ecosystem of software if one is already familiar/comfortable with it.

https://discoveryparts.com/products/...BoCQlQQAvD_BwE

For the budget conscious the now outdated "evo 4" (note the lack of an "S") are out there pretty cheap used as people are upgrading. It uses the previous generation software, but still gets you 1000Hz. My opinion these things are an unbeatable bargain if one wants to get into datalogging on a budget. I don't recall but I think the gps updates more slowly on the old evo4 which could be a problem if track mapping is really important to you.

https://www.ebay.com/itm/12730481203...hoCZSAQAvD_BwE