camman

s4 yes, s5 no

sft3303

Sweet, that's what I was thinking (the wiring diagram pointed that way, but I figured I'd ask). That's more wiring that can be taken out and obstructions removed from the intake piping. Awesome.

sft3303

Fuel pump issue figured out, I think. It was written on the webpage, seems like a silly mistake on my part to miss it in the first place. If this works, it should be:

1K-----Fuel Pump

That provides a ground to the stock fuel pump relay, and should do exactly what we need it to. Still digging for more, and will update as I find it out.

Aaron Cake

iTrader: (1)

Just remember Ohm's Law: V=IR. So you could Ohm out a harness and it will show perfectly under the microampres of current and low voltage from the meter. But suck 10A @ 12V through it and you'll have a voltage drop of 5V.

The MegaSquirt grounds the fuel pump output. Run switched IGN power to one side of the relay coil, connect the other side to the MegaSquirt output.

Ignition outputs are actually the same as most other engines with coils w/ignitors. Only difference is that the trailing has a toggle signal but this is fairly well documented in all the MS rotary writeups.

BAC, like the fuel pump relay and really most any other relay/solenoid in the car, is ground switched. One side of the BAC gets switched IGN 12V, other side connects to MS idle output.

This is a major guess for me not working with the PNP, but I'd say 4 cylinder wasted spark.

sft3303

Thanks a ton Aaroncake, I think I'm down to just the last parts of it now. I got all the fuel stuff sorted out, and the system primes just fine (well enough to point out my leaks at least)

I'm sitting in the car now on my laptop, and it won't start, but it's a flat battery, nothing else (or at least we can't be sure it's something else until the thing cranks...) My BAC was firing like crazy with the key in run (like a constant buzzing), even though the engine was not turning. Not quite sure what would cause this, but I'm fairly certain this is software related. It'll be unplugged for now, and I'll tune without it.

Edit:
Looking through the datalog, I was getting a steady 14 V at key on, and it only dropped to 12.6 volts while "cranking." Yet the starter never cranked over. The ecu doesn't touch the starter control circuit, does it? The wiring diagram has a wire coming off of the starter circuit going to the ecu, but it is in parallel to the starter, not in series, or controlling it.

muythaibxr

Those errors mean you have 2 features using the same output pin. Reloading the firmware will fix it until you break it again by causing 2 features to use the same output pin again.

Ken

sft3303

Duly noted. Thanks for the support! I think I found out later what I had actually done, as I was looking through my "bad" msq file that threw that error.

Still fighting the starter now though...

sft3303

The car is now cranking. Simple grounding strap was loose, from when I had re-installed the transmission. Everything under the car is buttoned up now.

Got it cranking now, but the car won't catch and start up. All my sensors are reading well, the CAS is syncing up just fine. Would an RPM below the RPM set in the cranking settings cause the car not to start? My datalog is showing 200 RPM for cranking, and I'm using Aaroncake's base settings (from his website post "How to Megasquirt Your 2nd Gen RX7") with recommended 300 RPM cranking speed. Any chance this would change anything?

I'm not sure what my spark outputs are doing yet, but I'm going to change this cranking rpm setting and see what happens from there.

I did flood out the rear rotor, should I pull fuel from the VE map or is there a cranking map I can pull fuel out of? On the plus side, I know my injectors are working, and I've got fuel.

EDIT:
Cranking rpm settings would do nothing. So nevermind. Still haven't got a running motor.

sft3303

Again, sorry for the multi-posts, but I wanted to update the status of the car, as I'm still a bit confused.

I didn't change the cranking rpm settings, as that shouldn't have done anything in the first place.

I am not getting spark. We checked with a timing light at the spark plug wires, with a volt meter at the leading coil signal wire, at the megasquirt jumper board.

So: the megasquirt is not sending a spark signal to the wiring harness. I measured this by placing a voltmeter between IGN 1 and GND at the megasquirt as the car was cranked, and there was no voltage signal generated.

I'm not entirely sure where to go from here. I'm fairly confident in the build of the megasquirt, but I'm not so sure in my ignition jumpers/software setup. Any pointers here would be great.

Again, sorry for just bloating up my own thread, but this you guys ( & the great guys over at diyautotune) are the best help out there, and I really appreciate what I've gotten to get this far. Thanks all.

SirCygnus

iTrader: (1)

Use test mode to check for spark output and check test mode for injector output.

sft3303

I've not heard of this before. Is there somewhere I can read more about this?

Edit: Megamanual. Duh.

What exactly does this let me do? The manual says:

"To enable coil testing:
- choose the desired output interval (the estimated RPM equivalent is shown below.)
- set Coil testing mode to One
- choose which coil to test e.g. CoilA
- choose your dwell time in milliseconds. Start low. e.g. 2.5ms.
- if your coils are powered via the fuel pump relay as is recommended, set the fuel pump to On.
- with valid settings entered, the Start button should be active - click it.
- when finished, click Stop "

What happens between start and stop? A spark at a given dwell time? Basically, what results will this give us, and what will those results tell us?

sft3303

Last update for a while, still no spark

Talked with Matt over at diyautotune.com and it looks like spark test mode is not available for the rotary motor yet.

Current status:
Had a couple of basic errors in my software setup (i.e., telling the software we were using a microsquirt diypnp module) and after that, I got a constant 5V signal with the key at "run" at the leading coil. Switched the spark setting to "going high/inverted" and the signal at the leading coil switched to a constant 0V.

After all of this, while cranking, I would get what appeared to be (on my analog multimeter) a constant 1/3V signal at the leading coil. This generated no spark, and the engine would not run.

Jumpering 5V from the VREF header on the DIYPNP board to the IGN 1 output, then removing the jumper would create a spark (this was tested with a spark plug pulled out of the engine).

I installed the stock ecu, and the car started, and ran (albeit poorly, but all the same, it started and would run well if you held the gas down and RPM's above 2000) so I'm fairly certain we've basically isolated the issue to inside the DIYPNP box.

Anyone have some insight? I'm lost.

sft3303

Update:

I've wired an LED inline with the spark output circuit, to get a glimpse into what's going on.

With rotary mode on, I either get a constant on light (spark set to "going low") or a constant light off (spark set to "going high/inverted"). This constant light is true whether or not the CAS is being spun.

If I turn rotary mode off (and change no other settings at all) then I start seeing what looks like a spark signal. A flashing light when the CAS is turned, and a constant light off when the CAS is stationary.

Any ideas folks?

SirCygnus

iTrader: (1)

Can you post up your pinout from the MS to the adapter board?

Im trying to do this on my side as well with a unit that i have.

ben@diyautotune

I assume this testing is being done off car, with a power supply and LED. It is not advisable to do this with your ignition hooked up as the "going low" setting is likely to damage your coils. The ignition output setting must be set to "going high/inverted" when the coils are hooked up. This is very important or damage will occur.

Using the LEDs as spark output testers is fine. A multimeter won't do a lot though as they generally can not respond quickly enough to the square wave, pulsed ignition output. The best and most correct tool for looking at the ignition output signal is an oscilloscope.

It sounds like you have a problem somewhere in your configuration, either in the project properties and/or in the ignition settings. As you mentioned that you had problems previously in the project config, I would start by reloading firmware to the DIYPNP, and *make sure* that you load firmware as a microsquirt. Then set up a new project, again in the project configuration set up the ECU as a microsquirt module. The procedure and correct settings are shown in detail at the tutorial at the DIYPNP website:

Firmware loading

Project Config

Installing firmware and setting up a new project provides a blank slate to work from, in case you are dragging in any old problems. Think about it like ripping off a band-aid, let's just get it over with.

From there, you'll need to get your ignition settings correct. What trigger wheel are you using -- factory FC style CAS? What coils are you using, stock FC or something else?

sft3303

The testing is being done on a bench, but the coils are not hooked up. I realize that damage could by done by leaving the "charge" signal (constant 5V) on, and as such made sure that this isn't happening.

I'm using stock FC ignition set up. So factory FC CAS, and stock FC coil setup (wasted spark leading coils, toggled trailing coils)

I'll try starting over with the firmware, perhaps I loaded ms2 firmware or something. I'll start over fresh and go from there.

SirCygnus- the pinout for the break-out-board is on page 1 I think, but it's missing ignition outputs iirc. I'll post up another pinout shortly.

Thanks so much for the help guys.

sft3303

So I reloaded the 3.2.4 firmware, started a brand new project, basically started all over, and it is still giving me the same problem. Basically works in every mode except what I need.

So it gives me spark in rotary "off", and in rotary "on" with RX8 mode enabled, but not in FD and FC mode with RX8 mode disabled.

Any thoughts?

SirCygnus

iTrader: (1)

send the msq to ForumHelp@DIYAutotune.com. ill be able to see whats going on.
Also, send me the pinout of what you got going on.

ben@diyautotune

As Peter said, I'd like to verify your .msq.

Also, please verify that you installed a 5V pullup on the WLED position (R2) for spark C logic signal.

In FC mode: Spark A is leading coil logic, Spark B is trailing coil logic, Spark C is coil select.
RX8 mode uses all 4 spark outputs for use with coil per plug.

The correct setup for a stock FC ignition system would be rotary on, FC mode enabled (FD mode and RX8 mode disabled).

sft3303

I'll send a copy to that address, do I need to send a copy to another address as well?

I've got the 51K pullup in the R2 position.

I'm aware of how the proper setup should be for my car, I'm just telling you what did produce a signal (as if it may eliminate circuit problems or something).

Currently, all of my testing with the LED producing spark signals has been from the spark A output. (Connecting the LED between spark A and GND)

If it helps at all, my testing setup is out of the car, on a bench, with nothing but a CAS, the Megasquirt diypnp v1.5, and my laptop. I'm away at school and the car is not near me currently. I will be able to go home this weekend, but just so we're all on the same page, I don't have coils connected, I'm not trying to start the car in FD mode or anything, I'm just kind of playing with things to see what makes the LED flash and what doesn't, and reporting that back to you. I'm not going to try to connect it to the car and start it (knowingly) in any other mode than what is prescribed for my vehicle (s5 n/a FC).

Thanks guys.

ben@diyautotune

Got your msq via email.

Under basic setup > tach/ignition settings, please change the number of coils from "single coil" to "wasted spark" under the pull down menu. Power cycle and re-test. I think that will fix it.

Also R2 should be either 100 ohm or 470 ohm, 51K is much too high of a value and wouldn't provide any significant pullup.

The 51k goes in R21, and you should have R20 and R39 unpopulated.

sft3303

Responded via email, but for those of you playing along at home:

I must have had r2 and r21 confused in my head when I said that. i have a 100 ohm resistor in the r2 position, and a 51k installed in the r21 position. r20 and r39 are unpopulated.

Changing to wasted spark did nothing, and I believe that setting was already set that way before I flashed the firmware and started over with a new project.

Just figured I'd copy a response here, in case someone with the same problem were to search for a solution.

sft3303

I wired up some more LED's today.

They show the signals at IG2 and WLD, and I get the trailing fire signal and trailing select toggle signal at their respective pins, but still nothing from IG1 unless Rx8 mode is engaged.

Just a thought, could I just use the signal from IG1 in RX8 mode to fire the leading coil, and just go without trailing coils for a while? Just so I can have my car back?

So basically, is the signal generated by IG1 & IG2 in RX8 mode the same as the signal generated by IG1 in FD/FC mode? Or is IG1 and IG2 separated by 180*? (like is this a true COP setup, or are the two leading plugs in RX8 mode still firing in a wasted spark manner?)

Just a thought, if it's dumb then I won't try it. But I'd appreciate some thoughts on that if I can get any.

SirCygnus

iTrader: (1)

you can run without trailing coils. just dont run boost.

denverstone

Just for those of us following along, can you share the email information if it is pertinent? I am building this kit as well but I am waiting to see if you can sort out your issue before I head down the same path.