MX-3.com

Inodoro what do you do for a living?

(sorry off topic)

you're not really understanding what I'm trying to say.

The neutral switch circuit, when not modified, is a ground input to the ECM. right?

so if you add a 12V ANYWHERE on that circuit, at some point, the ECM (expecting a gnd signal at some point, or no signal) will suddenly see +12v... do you see how this is bad?

Ryan wrote:The neutral switch circuit, when not modified, is a ground input to the ECM. right?

Correct.

Ryan wrote:so if you add a 12V ANYWHERE on that circuit, at some point, the ECM (expecting a gnd signal at some point, or no signal) will suddenly see +12v...

It won't, because the relay is a load. Just think about it. Does the ground side of your dash lights see 12v? No, because it's just that - a ground.

If you're concerned, you could put a diode in between the ECM and the relay...

Ryan wrote:

Where would you put the lead back to the ECM?

in between the neutral switch and the relay (where the line goes vertical...around there).

PoisonDrop wrote:Inodoro what do you do for a living?

(sorry off topic)

Whatever pays the bills.

However, I do see where you're going. I've been into electronics for the last 31 years (since '78), first as a hobby, and, since November '84, as a master electronics technician with a specialization in digital and industrial systems.
The only reason I don't work in electronics right now is because I'm not willing to waste 3 years of my life to get a UL certification.

PoisonDrop wrote:in between the neutral switch and the relay (where the line goes vertical...around there).

If you hook it up like that, the ECM will see +12v applied when the button is pushed.... Why is this okay? Sure, some of it is dropped over the relay load, but there is still a + applied voltage on a pin that is meant to receive ground...

its not like an applied voltage disappears or becomes impotent after the first load it sees.


and there is still a voltage on the ground side of dash lights. If there wasn't there would be no current...

When you said it that way it got me thinking...

If the neutral switch is NOT active, then you're right, the 12v will go to the ECM. If it's active (therefore grounded) the current should flow to ground. Add that diode and that should take care of that...

BTW Inodoro - your circuit would work as well, I'm not saying it won't. Without any testing, yours would be safer...

Either way I'm going to try this on mine tomorrow. I'm working on an engineering degree so this stuff really gets me going! I just have to figure this out now!

as I do this, I am attempting to get through my engineering homework too


If you add a diode, Poison's design would work. Inodoro's works, but woudl be safer if some resistances were matched up, if its at all possible. It would be perfect, if the neutral switch resistance was higher than the relay coil. Somehow I doubt it, though.


Poison: Even if the neutral switch was open, the ECM would still be seeing a positive voltage (although likely very small, but still the opposite charge). The ECM would act as a sort of parallel branch and take on some of the voltage... I just don't know whats inside the ecm behind that pin, so its impossible to tell what would happen....

Well I guess I was assuming that there would be a diode in the ECM on that pin. Or at least something to catch such a small voltage. But now that I think about it, maybe not, because with R&D, who need's a failsafe? It's not like Mazda WANTS us to be messing with this stuff...

as I do this, I am attempting to get through my engineering homework too

Engineering calculus...

God I start back on Monday...

Ryan wrote: If you add a diode, Poison's design would work. Inodoro's works, but woudl be safer if some resistances were matched up, if its at all possible. It would be perfect, if the neutral switch resistance was higher than the relay coil. Somehow I doubt it, though.

If you add a diode, it'd work, but I'd also add a resistor in parallel with the neutral switch (a big one, probably in the range of 10 M ohms or so), to make sure the voltage at the ECU stays low.

I don't know what you mean by matching resistances. The resistance of the switch will be much lower than the relay when the switch is closed, and infinite when it's open. What you need to be concerned with is the switch's current capacity. At 12V, a 100 ohm coil will draw 120 mA. I'm pretty sure the switch will have no problem with that load.

I was more concerned with the more delicate part, the ECM.

I imagine the neutral switch and the relay switch would show similar resistances, though.

PoisonDrop wrote:

as I do this, I am attempting to get through my engineering homework too

Engineering calculus...

God I start back on Monday...

CALCULUS!!!
Lucky you...!!
There's a part of electronics I don't miss...

Ryan wrote:I was more concerned with the more delicate part, the ECM.

I imagine the neutral switch and the relay switch would show similar resistances, though.

They will for the ECU. Remember digital inputs are normally very high impedance (typically 1 M ohm minimum), so if a switch is 2 ohms and the other is 500, it's still nothing in comparison. As long as you have a stable ground reference at the input, you're ok (and you could always put a small capacitor between the switch contacts, in case you have a bouncing problem).