Coil regulation related schematic diagrams and issues

[STC]

Ok. Indeed, "dwell time regulation" is incorrect - oily oil
Second. I think, I can say "dwell regulation" because dwell depends on voltage. Why not?
p.s. I need to clarify that because I want to fix wrong names of variables, defines, comments in source code.

[Fred]

Regulation is something you do to something that would naturally be out of control. Society, voltage supply, etc. Dwell is defined, not regulated. Set, not regulated. etc. You choose it from the outset, it doesn't come at you going "I AM 15 milli seconds damn it!" and you fight it off and reduce/regulate it to 5ms, you just choose 5ms up front. Make sense? Again give me some examples, make use of the code box on the forum, paste comments here, etc. Once I fix a few, you'll likely understand what the right direction for others is.

[STC]

Code: Select all

#define COIL_REGULATION

Code: Select all

#ifdef COIL_REGULATION
 volatile uint16_t cr_acc_time;       //!< accumulation time for coil requlation (timer's ticks)
 uint32_t acc_delay;                     //!< delay between last ignition and next accumulation

How about "dwell control"?

[Fred]

Yes, dwell control could be OK depending upon the context.

I don't understand, though. SECU-3 is (primarily) an ignition controller. The two jobs of such a device are timing the ignition event and ensuring the coil is charged correctly by dwelling it for an appropriate time for the voltage supply level currently available. In that case, when you undefine COIL_REGULATION, what exactly is SECU-3 doing? Logging RPM and MAP? Surely it should _always_ be controlling dwell?

[STC]

When I undefine COIL_REGULATION SECU-3 will generate fixed duty signal (some igniters accept such signal, e.g. igniters based on L497 IC). Another words, SECU-3 simulates Hall effect sensors in this mode . Also, some people use that mode for CDI
Ok. I will rename COIL_REGULATION to DWELL_CONTROL and so on

[Fred]

OK, in that case you should probably label the define as FIXED_DUTY_DWELL instead. That's just my opinion, though. I think code should tell you what it does with minimal comments :-)

[STC]

I tested following schema:

Сurrent limitting: