We have spark! (MegaSquirt)

[Gimpster]

Attached is my latest wiring diagram revision and my current tune. I also attached my latest datalog in the .zip if you're bored enough to look at it in MegaLogViewer (or whatever you like). You can see a few of my 1-2-3-4 pulls to 6200rpm.

Car's running pretty damn well. I'm getting to the point where I need to get on the dyno, which I plan to in the next month. I want to rent the dyno for the day and also put my truck on, it needs some tuning too. My local dynos a SuperFlow brake style, so that will make doing load cell tuning a little easier too.

I'm going to go ahead and call using the "smart" 6 COP igniter a success. If you look at my dwell settings, you'll see I have them fixed to 1ms. This is basically just enough dwell to fire them .. and since the igniter handles dwell on it's own, its all good. I was using fixed duty "Minimal for HEI4" but even that was too much dwell at idle. The igniter was getting HOTTTTTT!!! Let the car idle for like 30 mins last night and its cool as ice

So long story short, I'd call my megasquirt installation finished.

Here are pictures of my finished product in the trunk. Note that I made a big ole CPU heatsink into an igniter bracket with the tig this weekend. I figured it was a good idea. Pretty much the only thing left to do is encase the wiring harness adapter in potting compound, so its nice and secure (the base to the original ECU plug).

[singleslammer]

thats awesome man! you want to make me a setup?

[aler_ka]

just wodering, how much did u spend for the whole set up

[Gimpster]

If you include the new AEM wideband, $550 or so.

[aler_ka]

wow.. nice

[91mr2boy]

I have a quick question what is the "ignitor" that you are using?

Also, what is the COR on the diagram?

[Brad Bedell]

Circuit opening relay...

[Gimpster]

What he said Its the large black relay in the engine compt. fuse box.

Thats the 1995 Avalon igniter, which was my donor car. Except for some cosmetic differences (mounting holes) its the same as the COP bases JZ series igniters. I think all of the early 6 COP setups used it, but don't quote me on it. I feel dumb.. is it IGNITER or IGNITOR? I have a 9th grade Florida education, so bear with me folks.

[ZachAttack]

lol...english is a difficult language, even for natives


Gimpster, I'm considering going Megasquirt on my 3SGTE....on 1-10 how hard would you rate it for someone with moderate mechanical experience?

[Gimpster]

It would be a lot easier for a 3S owner like yourself. The 3S uses the 24/2 tooth style dizzy as I recall .. and there are a ton of writeups concerning that setup.

I'd give it a 5.

It wasn't terribly difficult if you've dealt with wiring, soldering and such. I'm not sure what your tuning experience is, but a good "basemap" to CAREFULLY start with might be one of the 4G63/Mitsu fuel & timing maps. The 4G63 is a very similar motor to the 3S. I've been trying to talk my shopmate into a MS, but he's got his heart set on that stock ecu solution that one dude is working on. I can't blame him though, nothing runs the car like an OEM ECU.

[Gimpster]

So hey everyone, I've been having some reset issues all the sudden. Lucky for me, thwam already worked through these issues here:

MSx/Extra EFI :: View topic - Constant resets under load, also while turning on headlights

Soooooooo I've added those filtering caps and we'll see if it gets any better. My resets were occuring during cruise mostly. Didn't have issues when I turned on the headlights or under acceleration like he did. They were very infrequent too.. like maybe once every hour or so.

Otherwise- no news is good news. It's running pretty decent, and I autoX'd the car this weekend. I don't have my tuning laptop here, but I'll share my latest wiring diagram and map here soon. I'm currently putting the finishing touches on getting the IAC (idle air controller) working. It's working pretty well, just gotta tweak the values a little more on warmup. Having the IAC is very nice. It idled fine, but took forever to warmup due to no cold idle-up. As usual, I'm using the stock toyota 3-wire IAC unit.

[Gimpster]

Updated wiring diagram and my latest calibration. BTW the firing order in the diagram is flip/flopped. I'll fix that in the next revision. Sorry about that!

I finally have my idle air controller handling my warmup idle, so that's pretty sweet. I need to get unlazy and setup my open-loop idle stuff next.

Cruise, idle and power AF is pretty much tuned too, I wish some of you guys were closer... I think you'd be shocked how well the car runs.

Adding the resistors in my previous post has solved all my reset issues. Also, I found out that having the MegaSquirt directly attached to the metal chassis creates ground loop problems. SOooo I went and found little rubber stand-off's and nylon bolts to hold the ECU to the bracket.

I'm about to start doing the full howto/writeup since the project is basically done.

[karl]

interesting. you have the center line to the IAC going to ground with the MS modulating the high side?

is that the way the V6 is wired stock? i know the 3S uses the center channel for 12V to the coils.

also, hwo are you makign your diagrams? is it just paint? i need to make some diagrams like that eventually

[Gimpster]

The V6 stock has ISO (open) and ISC (closed) going to the ECU being pulled low PWM and then just a +12V like pictured. As you noted the MS (my MS1 anyway) only has one PWM output for idle control so you I have to make it fight the close. Not the most efficient thing in the world, but it works! You don't need the closed if you only wanted to raise the idle.

I'm using photoshop to make the diagrams, only because I wanted layers for making changes easily. I'm not an art major, but I hope they're useful to someone.

[karl]

Originally Posted by Gimpster

The V6 stock has ISO (open) and ISC (closed) going to the ECU being pulled low PWM and then just a +12V like pictured. As you noted the MS (my MS1 anyway) only has one PWM output for idle control so you I have to make it fight the close. Not the most efficient thing in the world, but it works! You don't need the closed if you only wanted to raise the idle.

ahh i see it now. that's exactly how i'm doing it with the P72- i have the 12V feeding both coils with the RSC coil grounded fulltime. with the exception of it not idling higher when the motor is cold, it works great. you can see the IAC valve 'catching' the engine as the revs fall to idle, then letting it down to idle speed nice and easy. i'm actually pretty impressed with how well the stock MR2 idle control works with other ECUs

and more on topic i guess... i've checked the 3SGTE's igniter as well, since after readingi this i was a bit concerned about the fact that the honda ECU is trying to make the igniter (pretty sure it's igniter, not ignitor- i was confused for a while too ) drive the coils to about 3-4mS of dwell that it would be getting really hot, but it was almost ambient temperature when i checked it. i would think the only time it would get way too hot is if it's getting SO much dwell that it's going into current limiting mode and using the power transistors to limit current. seems at least in my case, the P72 wants to give it the exact correct amount of dwell. i'm guessing though that your COP units, having far less inductance than a standard coil, probably didn't require much in teh way of dwell and it was causing the igniter to go into current-limiting mode which is where all the heat came from.

i'm really fascinated by this ignition stuff if ya couldn't tell

[Gimpster]

Ever seen a Honda igniter? They are rinkydinky little things that are prone to failure... I would besuprised if they run much dwell from the factory. All that being said, there are so many Hondas out there maknig 500+ HP on bone stock igntion.

It's kinda weird, that's some of the stuff nobody has really cared to hack in the Honda ECU's.

1.) Ingition triggers, to use on different engines
2.) Ignition dwell

Both still pretty big unknowns to those guys, and they really don't care. It's kind of a shame since the Honda ECU would make a pretty good standalone if you could use it with different toothed wheel configs and all that jazz. I have actually asked JR (guy who wrote Neptune) to look into it, since hes kind of a buddy... but he's shown zero interest.

[karl]

yeah i was a little surprised by that too, considering how hot a lot of the DIS setups can fire. i'd think they'd be wanting to ditch the distributor.. that's on my list for the end of summer. the splitter circuit is amazingly simple, and i think i have a nifty little analog circuit that would allow me to reduce the dwell to run something like GM or the pantera EFI DIS coils. i wonder if it would be possible to run a sense resistor in the coil circuit and log coil charge current through the ECU (possibly with MS as well?) that would definitely make it easy to figure out where your dwell settings should be.

so calvin never looked into it with eCtune either? i guess i always figured that if the ECU knew which injector was supposed to fire, then it knew which plug was gonna fire right after that, and there's an individual cylinder fuel trim in the code, i'd be surprised if there wasn't an individual cylinder ignition trim as well. i'd think that's something they'd want to get figured out, since without it, you're only tuning the ignition to the worst cylinder.

[Gimpster]

I honestly just don't think those guys care about anyone wanting to use a Honda ECU on anything non-Honda. I doubt it would take much to nail it down, but there you have it. You do see the occasional COP used in the Honda world, but its usually ******/weak motorcycle coils rigged up in some manor.

I'll try to find the thread on a forum I bumped into, but there was some dude who made a sweet ignition coil and COP testing rig out of a megasquirt. He was using some super fancy scope though, but was testing all sorts of coils. He had some interesting data on various coils. One of the hotter coils he tested back then were the coils off an LS1.

Wow, arent we offtopic!