Why Rotax-Ducati have implemented such a complex kill magneto circuit in CDI modules?

Hi Juan, congratuations on you licence and Skyranger purchase. From what you say it sounds like you have some expertise in electronics.

I too have no idea why they have designed such an elaborate circuit to disable the ignition. I have seen circuits like this before and they tend to be designed by engineers who enjoy using the specific characteristics of components to design analogue circuits.

Personally I prefer to have more of a modular design where it can be clearly understood and there is less interaction between the various modules. Unfortunately I have too much experience of the design and its failure modes. In fact I have an old 4 pin module off a 912UL that now fails to start the engine. I have also had ones where the kill circuit has failed and it is no longer possible to turn off the engine without cutting off the fuel.

I much prefer the Ignitech modules, if I was designing one myself I would take the same approach that they have. I like the idea of the 12V helper circuit to power the ignition during cranking before switching over the Rotax charging coils to sustain the engine.

If you have read the forum posts about the ignition modules then you will have seen the name Mark Kyle, he claims to know someone who knows the design intent behind the Ducati kill circuit. You can find his details on one of the Australian ultralight forums if you want to ask him. I would be interested to know the answer too ;-)

Hi kevin. Thanks for your answer.
Efectively I have read several threads about CDI modules and  “reverse engineering” initiatives and related projects from you, M. Kyle, M. Miller and others.

I had a “difficult to stop” episode with my 912 in June and although finally it seemed to be a bad earthing connection problem, I had to research and learn to try to solve it. In fact, docs you uploaded to google drive were very helpful for me.

As you probably know, in Spain in summer it is very hot, CDI modules are negatively affected and there are many owners who have suffered problems with them. Needless to say, the cost of these modules are ridiculously and unjustifiably high.

From what I'm seeing, programable Ignitech's modules are pretty reputable and are an option for replacement.

Carmo repair services are other way but, after personally struggling for several hours trying to dig up the PCB of a bloken module, I really don't know how they can do the repair without breaking something else.

Another interesting option is M. Kyle's analog approach of replacing it with an improved module in its components. That is where I am seeing the possibility of eliminating the current kill system and replacing it with the traditional system “de toda la vida”.  It would be interesting to know Ignitech's implementation of the kill circuit. I think I don't risk much if I bet it doesn't have active semiconductors (MOSFETs, Thiristors, BJTs) 

As you mention, M. Kyle was active in this forum but a but not now. I will try to contact with him. 

I'm glad to hear that some of what I post on here has been of use to you, thanks.

I too would like to know how Carmo neatly remove the potting compound. I suspect that they have a template that allows them to cut the case in exactly the correct area of the faulty component. They could then use a small milling machine to remove as much as they can down to a safe depth and after that use a hot air tool and pick out the pieces. It might be possible to use a chemical to disolve the potting but it may affect the components or the surrounding potting making it difficult to repair.

My personal experience with the Carmo approach is that it is not reliable. It is always going to be difficult to repair a circuit board when you can only see a tiny window into it.

I haven't seen inside the Ignitech modules (I would like to) but I suspect that the kill switch input just sends a signal to the microcontroller which then just stops outputting the coil trigger signal. I doubt that there is any particular safety in this circuit, it just relies on the design being well tested and the manufacturing process being quality controlled.

Hello John.
My name is Alejandro Zapata and I am also in this forum. I think the thread is very interesting and I have transferred it too, although without your permission and I hope it does not bother you, to the Aerodisordered forum in Spain where there are great technicians where this topic is very interesting

I don’t see any issue in the “Kill” circuit as designed. It’s simple, passive, and more importantly, fail safe. The design simply prevents the ignition cap from charging when the kill wire is grounded, all the other components in the kill circuit form a low pass filter to prevent EMI noise from being generated along that wire to the ignition switch.

Internal operation of the ignition module  is generally poorly understood.  Maybe it’s time to shed some light on the subject.  Here is an explanation of the circuit function.