13Beast REW

You talking about the FD CAS? Its VR.

muythaibxr

Just use the built-in CAS ... You don't need to switch if it's already built in and on the e-shaft... then you don't even have gear-play to make your timing less accurate....

rmriggin

The pickups are mounted to the front on the outside of the front cover, and the FD engines have their mounts in the rear where as 1st gens have them on the front cover (why we would be using a GSL-SE front cover). Unless there is a way to modify the GSL-SE front cover to accept the FD pickups we are going to need to figure something else out. If I get the chance I will try and send some picks to show you what I mean.

Rex4Life

Sounds like you're gonna end up with a 2nd gen CAS. Nothing wrong with that, now that the 2nd gen CAS is mostly a plug n play with the new MSNSE code.

Two things I can tell you for sure:

1) The CAS will work with no problem on the GSL-SE front cover.

2) The timing mark on the GSL-SE front cover is in the same location as the S4 front cover.

I've never seen a FD front cover up close so I can't tell you much about differences with the FD.

FWIW, you should check very carefully the front covers for differences. I think there is at least one oil passage that goes thru the front cover.

HTH,

Scott

13Beast REW

Just got back from picking up the MS, so the project is officially started.

rmriggin

Will the 2nd gen cas work with the 3rd gen engine without having to swap e-shafts?

13Beast REW

Good question. I'd suggest calling someone like mazdatrix or speedsource on that one since they deal with both engines on a regular basis.

13Beast REW

Ken...
I noticed Ne- and G- both run into one wire on the FD ECU which by this post in the MS FAQ appears to be the same as the wiring for an S5. Is that correct? Basically what I'm wondering is if it actually matters that these wires stay coupled and just splice the single wire to the Ne- and G- pins on the MS or does the MS need them to be run individually to the MS CAS pins? Based on the post in the FAQ I've left them coupled and only have the Ne- pin on the MS connected right now. Do I need to splice the Ne- wire over to G?

Don

muythaibxr

well, it's best to separate them. Tofuball and I have it set up on his S5 so that they stay separate until after they get onto his daughterboard. If you link them together, you might get some interferance from the 24 tooth wheel on the 2 tooth input. (or in the FD case interferance from the 12 tooth on the 1 tooth).

13Beast REW

Redid the DB-37 and DB-9 connectors, both are completed now. I'll be splicing them into the harness tomorrow.

13Beast REW

Current status on the project is that all wires except for Fast Idle and Tach have been connected.

Wiring up the TPS was tricky for no reason since I just read the Service Manual Wiring Diagram kinda wrong. The C wire lead from the TPS (BR/B - Brown wire with Black Stripe) runs into the stock ECU at pin 4D. The ECU pinout chart in the wiring diagram lists 4D as a Ground for the ECU inputs but the wire itself does not run directly to ground. The BR/B wire also runs to the Intake Air Temp sensor, the Coolant Temp sensor, Metering Oil Pump, Fuel Temp sensor, and the MAP sensor. The BR/B wire is the ground for all of these sensors and runs back into the stock ECU at pin 4D and then is passed on to one of the stock ECU ground wires at 4A, 4B, and 4C (not sure which).

I've kept my stock ecu connections in place and simply spliced the MS DB-37 connector further up the wires from the stock connectors. When wiring up the TPS to the MS you'll have to also splice the BR/B (4D) wire into one of 4A/4B, or 4C. It doesn't matter which of 4A or 4B you use as you'll also notice that 4A and 4B are actually spliced together a little further up from the stock ECU connectors. I have 4D spliced into 4B at the ECU connect but I imagine that this may cause some issues with the stock ECU and should I need to use it again I'd remove this splice.

So at the current moment, my IAT, CLT and TPS signals are all reading as expected in MegaTune. The O2 sensor, all ignitor and injector wires are connected as well but I haven't been able to test them. I have no reason to believe that there'd be any issues with them though.

muythaibxr

for fast idle, what does the FD use? I'm using the BAC valve on tofuball's car... but doing that requires a transistor replacement that I might not have made for your setup.

Tach isn't going to be connected... don't connect it to anything... you're getting your tach signal from the CAS Ne/G wires.

Also, make sure that the shielding for the Ne/G wires is grounded, but only on the megasquirt side. If you ground it on both sides it won't do anything.

13Beast REW

The FD has an Idle Air Control Valve underneath the UIM.

Ne/G shield? I know what you're talking about but I figured I wouldn't have to worry about it since I'm splicing into the stock wiring right above the stock ECU connectors...

I'll post pics after connecting the IAC.

OH... in the FD FSM the IAC is listed in the wiring diagram as the ISC Solenoid Valve (Idle Speed Control) but in the rest of the manual its the IAC.

muythaibxr

is it a 2 wire PWM type or does it require a stepper motor...

If it has 2 wires and a spring inside we're good (this is most likely the case).

However, like I said before, I didn't swap in a ztx688 transistor for you I don't think... so if you come over this week, bring the MS and I'll take a look/add it if it's not there. The reason it's required is that it handles the higher amperage that the IAC/BAC requires. (As opposed to a relay, which is what the circuit was originally designed for with the pn2222a transistor).

The shield should definitely be grounded to the MS if you can... that'll get rid of any false triggers from the VR sensors and other such noise-related issues.

13Beast REW

ok update for tonight...

fast idle is still not connected until i verify that transistor.

Entered in the MegaTune settings and got everything setup and looking real nice. Problem now though, I'm not getting spark or fuel during cranking. When reseting the MS, I get the indicators showing the fuel pump is powered and I've checked with a spark tester that I get spark when booting the MS, but after that nothing.

muythaibxr

I'd have to see how you set everything up... Can you post an msq along with which version of msns-extra you're using, and your wheel decoder settings?

13Beast REW

I'll post it when I get home since I don't have access to my laptop right now. I'm using your settings from the FAQ right now and the default rotary settings from EasyTherm.

EDIT: LOL, damn it, now i'm not even sure what version of msns-e i'm using, i'll find out. its either 024s9 or 024s12

muythaibxr

Don't use 024s9 or 024s12... you need 025k or later to get proper rotary trailing support; the menus exist in 024s9/s12, but the code doesn't work.

Also we might have to do a little extra work on your car since it's an FD and you're the first to try getting an FD running on a megasquirt.

Also, are you showing 200-300 rpms on the tach in megasquirt? If you're not then the MS won't fire the injectors or spark... so that's the first thing I'd suspect....

13Beast REW

Yea I was just about to post about that. No the RPMs are not showing accurately. It's sporratic but the most common result is when starting to crank, RPMs will surge to 3k then drop down to maybe ~150-200 then straight down to 0.

EDIT: I've been looking over the schematics for the MS and the LM1815s since my last post and now I'm looking over the FD FSM for more info on the ignitors just trying to gain an idea of what I'll be looking for when I get home today.

muythaibxr

yeah, if you didn't ground the shield wire, you'll get tach spikes and stuff... (if you didn't ground it or if you grounded it on both ends).

It sounds like that's what's happening to you. You might also get weird behavior if the Ne and G are sharing the same negative wire... or even if the negative wires are just connected. Thats why I recommended keeping them separate.

First thing I'd do if I were you though is upgrade to a much newer firmware. I made a post in the FAQ yesterday saying that people need 025k or higher to run rotary trailing, but I guess I need to make it in bright big flaming letters

Oh, and since you already have msns-extra on your megasquirt, you can upgrade to a newer version without shorting the boot jumper.

Last, on your coolant/air temp sensors... did they read the same thing, or at least close? Was their reading close to the actual outside temperature? I'm asking because if you used the 2nd gen air temp curve for the air temp sensor, it won't read right... I believe the temperature curves for the air and coolant temp sensors should be the same on a 3rd gen.

13Beast REW

Got everything upgraded...

MegaTune - 225b627
MSnS-E - 026i

Still no reading on the tach. Had some other things to take care of so I didn't really get to dig down into the wiring last night though.

I'm a little confused as to how the problem with the CAS signal could be related to the Ne- and G- being spliced together or with the wire shielding being run to the body of the car rather than to the MS's ground wires. I did go ahead and separate the Ne- and G- signals, finding that the stock ecu is dependant on these two being spliced together. At the moment I have them in a manner that they can be quickly spliced or separated depending on wether I'm running the MS or stock ECU.

As far as the wire shielding goes, I was able to locate it. It was a little further up the wiring than I expected given that its still wrapped in the plastic tubing.

Its not fully registering why the excuse for the MS is that the shielding and merging of the Ne- and G- wires creates so much noise the signal is unusable when the stock ecu uses them just fine. The PFC and AEM do fine with them the way they are as well. Immediately the only thing I can think of would maybe be the rating of the input resistors

muythaibxr

There are a lot of things it could be... either way I think I'm going to have to look at it to see what's going on.

Ne- and G- being spliced together can cause little blips on the G wire every time an Ne tooth passes its sensor.

The stock ECU and other ECU's are fine with that signal because that's what they were designed for.... they probably use a different vr signal conditioner setup.

Anyway, we may also need to add load resistors on the Ne- and G- signals inside the MS... Either way we have to do some testing to get this working.... Like I said, you're the first person to try to get an FD working, so it's probably going to take some experimentation.

If you bring the MS and the FD over next week to my house (assuming you don't figure it out first)... I'm sure we'll have it figured out.

Ken

13Beast REW

Well the project is on hold until next week anyway now. Just found my grandmother died last night so I'll be out of town until Monday most likely.

muythaibxr

Sorry to hear that.

13Beast REW

Well... I'm back now so I should be getting back on this sometime today after I get some rest.