Is this pretty much how this works? Working on a distributor delete

Is this pretty much how this works? Working on a distributor delete.

bump

What car, how many cylinders, etc?

Anyone?

pls I don't want to toast a coil

E30 with an M20 straight 6.

Diagram is per single coil. Obviously this would be multiplied another 5 times.

Signal comes from the crank position sensor? Looks to me that every signal is a ground that energizes the primary coil which energizes the secondary to the spark plug. It all checks out for me, but I don’t know shit about how a BMW’s ignition system works.

Not going to claim to be the master of m20 engines - so expect some generalities.


1: You need a coil driver, I assume you're running COP or an ignitor/coil setup like GM did.

2) You can only delete the dist IF the cam/crank signals come from other sensors not housed within the dizzy (like hondas).

3) Yes, your basic wiring diagram is correct, power and ground, plus signal. Make DAMN sure the signals are entering the coils/driver in the correct pins. You can burn up a coil very quickly.

4) What I don't see is the most important part, the trigger or signal input. What is taking the mechanical rotation of the engine, and providing a signal for where the coil needs to fire?

Wasted spark would be much easier, FWIW

4) Cont.

Crank trigger wheel will have a gap for top dead center. You also need to determine if the ECU is triggered by the rising or falling edge of the tooth. ie: Crank output waveform is 0-5v hall effect. Does 5v or 0v trigger the TDC indication?

Just knowing crank position is not enough to run the coils, cam position will determine firing events, so you need to know what signal from the cam sensor can be interpreted as, say, TDC cyl 5. Cam/crank sync will determine correct ignition timing, so when number 5 is at TDC< you need to know what each sensor will be reading so the computer can fire number 5 when those inputs are met.


What is your ECU solution for this? Piggy back signals into a different coil/driver and use wasted spark? Or actual sequential ignition?

>4) What I don't see is the most important part, the trigger or signal input. What is taking the mechanical rotation of the engine, and providing a signal for where the coil needs to fire?


Not op, but there are two ways the M20 got its signal, earlier models had two CrankPS in the rear running off the flywheel and the later models had just one running of harmonic balancer in the front.


Honestly OP I think its shitty Idea, the bosch Motronic system at that time is very limited with what you can do with it.

Thanks for the replies.

I've got the CPS figured out. It's just a VR sensor on a tooth wheel. The thing I have marked SIGNAL is the signal from the ECU to fire the coil. I'm also installing a standalone engine management system, so that's handled.

I'm throwing together a little test board with an Arduino and a MOSFET right now that will pulse the coil for 3 milliseconds at a rate that I'll set (probably something low like 50 times a second, which equates to the engine running at 100 rpm, so nothing crazy). I'll have it firing through a spark plug and hopefully I don't either explode my microcontroller, toast the coil, shrek my power supply, or kill myself. High voltage scares the shit out of me.

I've been gone because I had to go buy a MOSFET to do this.

First I have to eat dinner and relearn how to program an Arduino.

>Motronic is limited
Yes it is! It's terrible. Which is why I'm going to run Megasquirt.

Fun fact though: the toothed harmonic balancer from the later cars bolts straight onto the crankshaft of the early cars that had the CPS in the bellhousing. It even has the TDC reference pin and the two holes for the sensor bracket. Weird, right?

BMW was probably limited with what it could export to the states, and typically whatever we got was a year or more behind what Europe got.

Why are you trying to delete the distributor?

t. dumb niglet

A bunch of reasons. It's kind of like "well while I'm at it" from running a standalone. I don't like the idea of wiring around high voltage, and I don't like needing spark plug wires. Distributor caps and rotors are also kind of imprecise by nature, imo. They wear out, they break.

If I use a coil on plug setup, there's only 12 volts running to each cylinder, which is neat. It also makes it much easier to play with spark timing and stuff.

My suggestion is an optoisolator.
If you are set on MOSFET make sure it has a good heatsink, and have a backup 'duino board at all times.

Ah, standalone, I have an MS3 myself.

No need for a test rig, MS3 (and I think MS2) have a spark test output mode, 1 pulse per second. MS will handle the trigger, firing order, etc.


FWIW, a direct shock from a coil (12000v or so), isn't going to seriously injure you, since there is so little amperage behind it. It will startle you, and can mess your heart rate up for 10-15 min. This is from right off a coil, ground path across the chest. With just holding the wire and letting it arc, you'll feel almost nothing.

There's only 12v going into the coil for the dizzy too. All you're doing is removing the spark plug wires, and giving each cyl a coil.


COP and true mechanic/vacuum advance dizzys with stand alone coils are not really very different systems.

>It will startle you

Getting soccer kicked to the shoulder would be startling, yea. I was sore for a full day.

Y- you g-got me g-g-good uncle mike, heh hehh... ow.... :(

>My suggestion is an optoisolator.
>If you are set on MOSFET make sure it has a good heatsink, and have a backup 'duino board at all times.

In either case an IGBT is a much more robust choice than MOSFET, less likely to Kaboom. You don't want that magic smoke getting out now, right?

I don't have one on hand, unfortunately.

I need to wire up a coil driver box anyway. I probably will use IGBTs in the final installation just like says. Gotta keep the ghost trapped.

That's why there's 7 high voltage wires for a 6 cylinder engine, right? The distributor cap has a hard line to the coil secondary, at least in my car.

I just finished building the test pulse generator. I'm terrified to hook the coil up to it. High voltage really, really scares me.

It doesn't seem to work. I might have the MOSFET hooked up wrong.

I'm using a computer power supply and some pretty thin wire though. Maybe it isn't able to charge the coil sufficiently because it can't provide enough current?

I'm out of ideas.

I'm dumb as fuck. It's because I'm firing it with 5 volts. Rewiring.

Exactly. One feed from coil to dizzy. Then the dizzy spreads out the voltage to each cyl as needed.

IT WORKS