There are 2 possible scenarios that could be being discussed here ..........

1)
You want to make a Diagnostic-Tool for finding a poor Ignition-Pulse that may be causing
problems with the performance of the Ignition-System.
OR,
2)
You are trying to duplicate the output of the Trigger-Coil / Pickup-Coil,
so that You can perform experiments on the Ignition-System without the Engine running.

If this is the goal, this project will be almost impossible to achieve,
and will have to be "adjusted" or "tuned" to match the particular characteristics of the
Model of Engine, and Model of Ignition-System, being experimented with.

These types of Small-Engine-Ignition-Systems usually operate on very similar principles,
but, at the same time, each design is also unique, or special, in some way.

What is the overall goal that You are trying to accomplish with your project ?
.
.
.

 

This is scenario 2. Simulate the signal; but I think that it should not be perfectly equal to the wave, because I think it is enough if it is a pulse, the electronic switch should work by edge detection or a threshold.
I will upload a diagram of an analog CDI, but there are other digital ones

 

This is the first idea I had or have to do before thinking about maintaining a constant useful cycle courtesy of another colleague. Which idea will keep the useful cycle more stable when varying the frequency, through monostable or through pwm?

 

Sorry. My conection!

 

OK, things are starting to become more clear now. ................

If You are trying to produce more Power from the Engine, You are wasting your time.
Get a bigger Engine, this is the only practical way to produce more Power.
Otherwise,
You will need to have a very expensive Chassis-Dynamometer to actually MEASURE THE CHANGES.
Guessing that some change is working is just that ...... Guessing.
And, You may damage some Engines by making them go into a state of DETONATION.

The best way to increase Power on a 2-Stroke-Engine is by designing a Custom-Expansion-Chamber-Exhaust.
Adjusting the Ignition-Advance-Timing is a very tricky business,
You can reduce Power just as easily as You can increase Power,
but any increase in Power will be very small.

It is entirely possible to make a Circuit that will duplicate the Signal from the Trigger-Coil / Pickup-Coil,
but it will be just a waste of time, so I'm not willing to take the time to provide it for free.
Learn how to design an Expansion-Chamber-Exhaust-System.
.
.
.

 

Thank you. But the main objective is not to increase power and much less to decrease it. It is knowing the advance curve of a CDI to know the curve that that motorcycle has. First, know possible replacements and second, if it can be used for someone else. Most are not 2t engines.

 

Here in Cuba many have been imported, mainly Suzuki GN125 and copies of Honda Cg125. Both 4 stroke

 

You will be able to appreciate
that investigating various ways of doing something for me has the main objective of learning.

 

This project is too complex, and has too many ways that it can deliver a very inaccurate result.

It should be designed using a Micro-Processor,
which I believe You said earlier that You did not want to use.

Different Engine designs will produce different Waveform-Characteristics from their Ignition-Trigger-Coils.
The different Trigger-Coil-designs may also may vary the Ignition-Timing with changes in RPM,
as well as the Ignition-Module changing the Ignition-Timing with changes in RPM.

This is not a simple problem.
.
.
.

 

here is the image from link
R1 is used t ocontrol the frequency.
R5 is used to control duty cycle.
both controls are independent of each other.

View attachment 327087

Certainly this scheme will work.
I visualize a digital scheme that can hold the duty cycle constant, settable in 1% increments, over a very wide frequency range. It is settable from 1% to 99% by means of DIP SWITCHES. That is an alternative. Totally digital.

 

The digital scheme would use an oscillator at 100 times the pulse rate, and a decimal counter to select the percent to rest the PWM pulse.

But now we have learned that the actual goal is much different.
This is another case of an individual having decided a solution that may not be correct and then asking us to make it work.

 

Friends, I'm sorry, I don't want you to think that about me, that's why I explained to you the problem that gave rise to this idea. I am open to hearing other ideas. I repeat that learning is also objective. I have less mastery of programming but it is the best, for example I did not know that it is achieved with labview. Also excuse my ignorance for thinking that the waveform for this case does not matter much, since I think that this case senses the frequency and time (sweep angle) between the edges, zero crossing of both opposite pulses, also this signal It is not like the case of cars that synchronize with another camshaft signal.
Do you think that with the astable and then a monostable like the design I published is more stable in practice? I was studying 4538, how accurate will it be with respect to 74123 or 555?

 

This is NOT a DIY project.

Use only the Ignition-Module that was designed for the particular Engine in question.
They are dirt-cheap.
.
.
.

 

This is NOT a DIY project.

Use only the Ignition-Module that was designed for the particular Engine in question.
They are dirt-cheap.
.
.
.

okay thanks.

 

This is NOT a DIY project.

Use only the Ignition-Module that was designed for the particular Engine in question.
They are dirt-cheap.
.
.
.

I have not seen any ECM ignition module at anything close to dirt cheap, at least not one that would last more than a week.

And now that I understand that the TS is wanting to research the programming of the electronic portion, rather than the actual operation of an engine, I suggest that using a signal generator and the digital scheme will provide more accurate data by means of repeatable pulse creation.

 

What is TS. Sobre!!

 

Sorry

 

Even if I can solve the simulator, I take it as a school task to learn. That's why I'm looking for several ways, each one will have to be studied and even though I already know that the best way is through software and programming that I will also study with labview, various doubts arose, such as the precision of integrated ones for monostable such as 74123, 4538 and 555 .

 

Your ideas have been very useful. For the moment I will do the 33 variant although perhaps with the ramp generator with operational or unijunction transistor, I will try to use operational from the LT or LF series. I will always check the duty cycle and correct it if necessary.

 

I have already tried with the type of wave that circuit 44 outputs and it works, but with another circuit that only makes it work in a range where the cycle is real for the CDI.