JustinL

I got my Megasquirt2 system up and running controlling fuel only. I'm using the stock wiring for the most part. Here's what I did for anyone who wants to try it and for my reference as well.

1.Throttle position sensor (TPS) Pin 22 tap and supply pin 21, as previously mentioned, works perfectly with Megasquirt.

2. Coolant temp works DME Pin 13 to megasquirt. The resistance needs to be adjusted accordingly in Megatune.

3. Intake air temperature sensor (IAT) is connected to the port previously used by the stock idle control valve. The open line to this will have a little air filter connected. It would be better to have a new port welded, but I'm having trouble finding someone who will weld aluminum. Wires will need to be routed under the intake manifold to the passenger footwell.

4. I'm using a PLX m300 wideband that works flawlesly with the MS.

5. Triggering is done with pin21 from the dme (tach signal). I found that the MS draws too much of the signal and doesn't allow the DME/KLR any signal left to process the rest of the ignition. There was no tach jumping on the dash and no spark to the coil. The solution was to put a 1.2k ohm 1/4 watt resistor in the line to the megasquirt. With the resistor in place, both devices have functioning tachs.

6. I decided for simplicity sake to reuse my ballast resistors and set the MS accordingly with my low impedance 55lbhr injectors. I'm using spade connectors in case I want to switch them out in the future. MS calculated a nice VE table for my setup and it's running well untuned and just a little rich now. The next step is retuning for a more exact fuel table.

7. The DME needs a load signal, which is usually derived from the air flow meter (AFM), to calculate spark advance. I guess any type of MAF or AFM would do the trick as long as it output a 0-5V signal. This signal goes to Air flow sensor Pin 9 on the DME. I'm reusing my old guru racing Link AFM to generate a load signal. Hopefully someone smarter than me can figure out a way to get around the need for this. The MS can also use a MAF signal to generate the fueling characteristics, so you could plug in a generic MAF and be off and running.

That is how I got my car running with it, I do not make any claims that this will work for everyone. YMMV!

If anyone else gives it a go and wants my settings files or tables, I would be happy to share; just drop me a pm.

Again, Big thanks to AlexE (ae1969) for his expertise and the laptop he generously loaned me.

Cheers,

Justin

mark944turbo

Cool!

I am still waiting for the next megasquirt to come out with fuel and spark control and lots of extra inputs and outputs. The specs on it sound great.

JustinL

Yes, once the router board and GPIO are finished it will be a mighty system. Until then it's just a nice upgrade from the link MAP kit. I just got it running tonight and tomorrow I'll be trying out some higher boost zones hopefully. I also need to clean up the current rats nest of wires in my passenger footwell and tuck everything into place.

IanS

Congratulations. Lots of progress lately.

I believe most cars have an IAT sensor right after the throttle body? I would think you could drill and tap the entrance to the intake manifold and add one there? My concern about the ICV line is that it might not get as accurate temp reading, since its not in the main airflow?

Does that mean you've deleted the ICV? Is that working out well with megasquirt?

JustinL

Ian,

I agree, The placement for the IAT sensor is not optimal, but the sensor has some reach and gets just into the airflow. I would like to reposition the sensor in the future to get better readings, but it would certainly require drilling and tapping the manifold or welding an aluminum bung on to the hard pipe. I believe the stock pipes are steel, and would be dead easy to get a bung welded on.

You're right, I have had my ICV deleted for some time. It can be a pain to keep the car running for the first minute as you have to keep the throttle slightly open, but once it's warm, the idle is rock solid. I originally did this to help with eliminating the hesitation in with the Guru MAP kit. With a MAP sensor style though, you can take air from anywhere for the ICV. No need to worry about metering it, just filtering it.

Justin

Perry 951

Same here!!!!!! Tried to get a MS1 to work well with my dad's project ('83 911 SC, 3.6 big bore on the 3.0 block, trick heads, Electromotive timing, and a intercooled Whipplecharger ) It hasn't run in about 3 years, and it's been a project for 13.

I'll also get the new MS for my 951, down the road.

IanS

Justin, am I correct in saying that we have a 135 tooth crank wheel measured by the speed sensor and a solid wheel with a single missing tooth measured by the reference sensor?

FYI, the new MS2 beta code will decode a 135 tooth crank wheel, but I'm not sure if it has second wheel support yet. But I wanted to make sure I knew what kind of second wheel we had before asking.

The new beta code also has knock detection, so if they had 2nd wheel decoding, I believe that would make the MS2 a fully plug and play ECU replacement, minus boost control.

Yes, I've now spent way too much time reading the Megasquirt boards. I even looked at the C code. Its pretty clean. I might even be able to add features if I knew anything about the hardware.

JustinL

We have 132 teeth on the flywheel and no missing teeth. The reference sensor is how the DME knows the position of the engine. Essentialy it's a dual wheel setup and would need to be decoded appropriately with one wheel having 132 teeth and the other having 1. I think once the router board is built, plug and play will be much closer to reality. Another option is to use the the current software with stock wiring and build a circuit that would flatten a couple teeth every time the reference sensor came past. This would essentially create a 132-2 wheel.


This is a good time for an update as well. I've been running the Megasquirt setup for a few weeks now and it's GREAT. Very easy to tune and drivability issues from the link MAP kit are gone. No more hesitation and rock solid A/F ratios. The limiting factor now is my tuning skill

The current issue Alex and I are both having is hot starts. I can't imagine that it will take to long to resolve that though.


Justin

(Go OILERS!)

BC

How bad was the innitial tune? How far off on fueling where you?

JustinL

MS2 uses engine size, injector size, peak tq, and peak hp to generate a VE table. It's not surprising that this 'guess' isn't especially acurate, but surprisingly everything ran pretty well right away. It went quite rich on boost (below 9:1) and I let it autotune all the zones in part throttle. Now it's running 11.4-11.7:1 under boost. I would say I had to scale the fuel back 10% across the rev range on boost.

IanS

Of course, I knew it was 44 x 3, but apparently I can't do simple math anymore

Sounds like the wheel setup is basically what I thought. If someone ported the MSnS/Extra 2nd wheel decoder(http://megasquirt.sourceforge.net/extra/wheel2.html) to MS2, it should work nicely.

The idea of doing the 2 tooth subtraction in hardware is an interesting one, but I don't know enough to design that myself.

The router board sounds promising, but I'm wise enough to know that you're better off working with existing hardware and slight modifications to existing software than waiting for the next greatest thing.

I'd be happy using MS2 for fuel only for a while, but I'd be stuck with the AFM until I could figure out how to generate a load signal for the DME or decode the flywheel and add spark control. Right now, flywheel decoding looks slightly easier and more useful than trying to reverse engineer an AFM signal.

BC

I currently have sitting in a box:

A V2 MS1 ECU
A full wiring loom for MS1 (extra)
That large patch cable
The relay box properly jumperd for MS1 (extra)
Two EDIS coils and related Shtuff for running MS+EDIS from extra

A flywheel that I made a custom 36-1 toothed ring for, and has fastened to it, and balanced for my engine build.

What keeps me from smacking everything together (besides time) is knowing whether or not Breaking in the motor on a NEW tune is going to be healthy for this pretty expensive motor.

I could use MS2, by just adding that second board, but then I lose, I think, that wheels decoder that would allow me to NOT use the EDIS -add on, and just run coils from the VB921s (that I haven't bought, nor would know how to install)

JustinL

It sounds like you've done the hard parts allready. Check out Alex's setup; he's running exactly what you propose to run. You can indeed use MS2 with EDIS and it works very nicely. http://forums.pelicanparts.com/showt...hreadid=286145

Make sure to read the earlier parts as well.

I've wondered about running an S2 or 968 flywheel with the 60-2 trigger wheel already installed.

Some adventurous soul with an S2 or a 968 should give MS a shot and maybe we can get a plug and play system going.

theedge

Those things are about $1k though...

BC

I was hoping to run the coils with no EDIS system. There are reports of the EDIS module cutting out at 6000rpm.