O R X O N

Hi guys. I have an absolutely batshit perplexing issue.
I have an MS3 + MS3x and an 05 NB2.
My MS3 will not power on.

I followed the instructions on Trubokitty to a T and have the pullups, the jumpers, the works. On a test bench (my desk + 12v Jumpstarter pack) the MS fires up - I supplied power to 4AF and Ground to pin 3B so I could flash the firmware on the site and upload the tune settings. I didn't want any of my Mustang outputs firing or acting weird. The positives and grounds are good, and I jumped the IACV pins as well on the back of the mainboard's DB37 sockets. I can power the MS from the MS3x injector ground as well so that's good. Power delivery appears to be solid on my wiring. I used this MS3 on a v6 Mustang and it ran and ... sorta drived. It was just for learning. A lot of the stuff is the same for the Miata (Ex VR conditioning)

Pin 4AF is listed as either red/white or black in the 05 FSM. For me, it's black. When I have Key On/Engine Off, pin 4AF to Ground is a 0v difference.
The 05 factory service manual shows pin 4AF going all the way to a positive. My 10A ROOM fuse is not blown. I can put the stock ECU back in and drive away in a matter of five minutes.

Any pointers? Something isn't adding up here.

O R X O N

Update: The FSM shows 4AF gets it's power through the main relay, and the main relay I do not hear clicking on. The main relay is triggered by NB2 pin 3H. I will report back after testing this.

O R X O N

Got farther. Trubokitty site shows 4AF for MS power; which is incorrect. 4AF is switched from the main relay.
Pin 4S is a power source (not used via Trubokitty) for Run/Start [4AG, correctly not used, is an always-on 12v source for DTCs]

MS will need to switch on pin 3H (not used via Trubokitty) to turn on 4AF (not necessarily used) as well as the injectors.
This doesn't just require changes to wiring diagrams, the tune settings as well.

I'll work on this later. It's 4am and I work in a few hours.
I'm watching you, Braineack.
In before 18psi. Goodnight all.

Braineack

iTrader: (62)

Lol. How about you look at the MSM documents?

O R X O N

Hey thanks, man.

I'm not on an MSM. I wish. Regardless. Because on the MSM setup the Nitrous pin used for the main relay is used for my VVT control.
Then my Knock sensor pin is 4M - which Trubokitty has correct for an 05, is not the same as an MSM.

Not being a dick, thank a ton for setting up a site like this. Lest I have to hunt down **** for a 99 v6 Mustang (read: there is virtually nothing. Because v6.) and reverse engineer PWM fuel pump control.

But for an 05, the "01-5" documentation is simply not correct. Has nobody ever brought this up?

If you want I'll shoot you updated HTML/Bootstrap and a tune file when I have a setup that is turn-key.

Braineack

iTrader: (62)

i built that in reactjs, that diagram is populated from a json file.

There had been rumblings about the 03-05 being different than the 01-02, but no, this has never come up before.

I found this diagram: http://www.mellens.net/mazda/Mazda-M...005/wiring.pdf

and i looks like it matches your observations, so I can update the documentation accordingly.

O R X O N

That's awesome. I work for a hosting company, but I do server side ****. Think MSP + VMware.

When I get it to boot up I'll test all the functions and make sure nothing else is different if you haven't already checked every-single-damn-pin on the 05 FSM lol.

Braineack

iTrader: (62)

yeah, I populate those digrams with this:

{id:'2A',wirecolor:'INJ A',function:'INJ A', mscolor:'W', pin:'X -19', output:'(CLY 1)', squarecolor:'filled'},
{id:'2B',wirecolor:'FAN',function:'FAN', mscolor:'WB', pin:'X - 1', output:'INJ G', squarecolor:'WB BTXT'},
{id:'2C',wirecolor:'A/CFAN',function:'A/C FAN', mscolor:'WG', pin:'X - 4', output:'INJ F', squarecolor:'filled'},
{id:'2D',wirecolor:'INJ D',function:'INJ D', mscolor:'WP', pin:'X -10', output:'(CLY 2)', squarecolor:'filled'},

all those buttons do it load the connector component and links to each stubbed db file.

Braineack

iTrader: (62)

after reviewing, i really see a subtle difference in the power handling between 01-02 and 03-05.

You'd simply need to add nitrous 2 to 3H to activate the main relay. then move power from 4AF to 4S. everything else is the same.

O R X O N

Oh Mazda.
Give me like three hours to get off work, get home, and re-adjust my harness. I'll jump for joy shortly.

Braineack

iTrader: (62)

well, not nitrous2, since that's used to control vtcs. you may have to add an output circuit on the mainboard and use that -- one of the LEDs.

unless youre not going to use BOOST for ebc, then that one would work.

O R X O N

I will not be using EBC, I'm still N/A - for now....

Would there be any ill effect of keeping the tumble control open even when in low RPM states?

Braineack

iTrader: (62)

I added it to my site. forcing VTCS open won't negatively impact anything -- its supposed to help with cold idle only

O R X O N

So, an update, got the ol' floating dimly-lit status lights on my dash.
When I plugged in the first connector, whew lads, the main relay went ******* bonkers. EBC was turning my main relay off and on. This was with the key not inserted btw.
Plugging in only the big connector with just power, I got proper key behavior and no floating lights; used this to change my EBC setting.
From here, when I plug in either the medium or small connectors, even with key off, things are happening in the bay but I can't pinpoint what.
The Megasquirt is not powering on - or at least not communicating. So something is fucky and backfeeding, somewhere. Also the immobilizer light is screaming at me even with key off

I haven't got my wideband or MAP tube set up, so for ***** and giggles after testing all of my IO in Tuner Studio, I fired the engine up on Alpha-N. WHEW! Pep in the step, ****** revs about as bad as it stunk.

Excited to get this finished.

O R X O N

Cut the BOOST - 3H (main relay) wire.
Those dim lights were a dead giveaway of a coil being energized.

When I pull 3H from the MS BOOST line, everything shuts off properly.
Reconnect 3H to MS boost and I get the float effect.

Now to find a suitable MS input that won't backfeed. If that's possible. Am I wrong for thinking a diode won't work here given that current is sinking INTO it when off, and has to be sunk into it to energize the relay?

Braineack

iTrader: (62)

curious. the inline diode might work, to prevent the 12v from the ENGINE Fuse to reach the MS output through the coil. That flyback mod on the expander connects that output directly back to s12 which could just enough to power things in the MS itself. I used to do that flyback mod slightly differently (with diodes) -- so might be why I haven't run into this before.

O R X O N

Either I'm going to relay the relay or find an opto-isolator.

.... Or steal one of the FETs from the MS Mainboard.

Braineack

iTrader: (62)

do a quick wire swap and try using an injector output instead, those are fets. see how that reacts.

I used to do flyback on the other side of those diodes pictured above, and each one through a diode to 12v. But documentation says do a single wire from one diode to s12, so that would be the only difference -- i really do believe that's the issue.

O R X O N

Well that'll do it.
Boost is controlling the charge light
Inj E is controlling the main relay.
Works perfectly.

Braineack

iTrader: (62)

nice! I'll update my docs.