[Antonio]

I am preparing to move a 6 cylinder engine to speeduino.

The ignition is coil on plug.

I want to run sequential ignition and injection, because it seems like the most tunable configuration, but I read that Arduino Mega does not provide enough timers to be able to do that, so I can use instead STM32 or Teensy.

Which one will be more easy to work with? Would it be Teensy, since Dropbear uses it, or it is easier to use STM32?

I can play with the firmware as I am a software developer, and I have some experience with electronics.

Thanks

[jonbill]

Out of interest, what have you seen that makes you believe sequential to be the more tunable configuration?
I think for the purposes of most easily having 6 channel sequential, it's not really a question of choosing between Teensy and STM32, its a question of which speeduino board that supports 6 channels do you fancy?

[Antonio]

If using wasted spark for ignition, it makes the coils "weaker" as on high RPMs they will not get so much time to "charge" and the voltage will not be so high, so you can't keep a bigger spark for a higher duration.

Regarding the injection, as I understand pairing injectors will result in both injectors opening at once, which is optimal for one cylinder, but suboptimal for another, because the fuel will rest on top of the valve until the valve is open.

So if I can control all injectors and sparks individually, it will result in better performance.
I don't go into this project chasing performance as ultimate goal, but I want to do things right, and it feels more "correct" to try for the best configuration from the start. I say this based on assumptions, as I don't have real experience on what is best.



Regarding the speeduino boards, I know about the V0.3, V0.4 and Dropbear. On the wiki there is info about 3rd party boards https://wiki.speeduino.com/en/3rd_party but I guess I have to search on the forum for more information about boards that support 6 + 6 channels.

I asked about the microcontroller, because I plan to start with the microcontroller and the firmware, and to add pieces step by step simulating the inputs, to make sure all parts work individually. I already tested controlling the injectors and coils with an arduino, but now I need to put together a lot of pieces, as I have an old BMW M20 engine and it will require some modifications, and I feel more confortable testing each piece individually in an incremental fashion.

[PSIG]

You have made several assumptions, many based on internet rumor. While there are exceptions to every rule:

If using wasted spark for ignition, it makes the coils "weaker" as on high RPMs they will not get so much time to "charge" and the voltage will not be so high, so you can't keep a bigger spark for a higher duration.

Wasted spark still has more than enough time to dwell for full-energy sparks. It is not a concern. The number of cylinders is unimportant, as only the pair of cylinders and RPM are involved. For example, a 4-stroke-per-cycle engine at 8000 rpm takes 15ms for one full cycle, firing the WS coil twice, each 7.5ms. Lots of time for both dwell and spark burn for even slower coils. Obviously, at perhaps 12000 rpm you would begin looking at quicker WS coils, or move to coil-per-cylinder (CNP or COP).

Regarding the injection, as I understand pairing injectors will result in both injectors opening at once, which is optimal for one cylinder, but suboptimal for another, because the fuel will rest on top of the valve until the valve is open.

This is true, but can have multiple advantages, and is the most commonly-used injection mode. Modern ECMs today often use two injections per-cylinder, per-cycle, for the benefits. Single-injection full-sequential can be beneficial for per-cylinder tuning, but you need to have the equipment and knowledge to use that feature effectively. Racing teams and shops do, but few others.

This means the project goals should guide what mode to use, without assuming one is better than another, or why. Weigh the benefits and detriments for your project, and pick the better fit. If you want to do it "right", then that's right.

So if I can control all injectors and sparks individually, it will result in better performance.

Again, an assumption, without basis for the judgment.

I don't go into this project chasing performance as ultimate goal, but I want to do things right, and it feels more "correct" to try for the best configuration from the start. I say this based on assumptions, as I don't have real experience on what is best.

Thank you for recognizing the assumptions. Try to go by facts, not feels. This will become very important when you get to running and tuning your engine.

I asked about the microcontroller, because I plan to start with the microcontroller and the firmware, and to add pieces step by step simulating the inputs, to make sure all parts work individually.

This is usually easiest by selecting the board that fits your project best, then assembling the circuits you need and test them. It greatly simplifies and speeds the process. Just suggestions.