Braineack

iTrader: (62)

weak.

Save your MSQ.

Go here and download this utility to reflash the MS:
http://www.megamanual.com/files/soft...ll_Install.zip

unplug your igniter/coils. You must do this else risk damaging them, and by risk I mean you will.

open Easy Therm. Select Custom on the side and and make sure the correct com port is selected.

download this:
http://www.boostedmiata.com/MS/msq/f...lone_noafm.s19

Power up your MS (key to ON) and click Download to MS; select the 10g_16L_standalone_noafm.s19 file you just downloaded from my site.

Ignore any prompts, just click yes. Don't touch anything while it's flashing else it will freeze and then you'll have to open the MS and install a boot jumper the next time you try it. You might need to this time if it's not communicating. you must open the case and install a little paperclip in the two holes where it says boot. then try flashing again.

It will load around 1775 lines. Once complete, close the program and re-cycle power to the MS.

Open MT and open the last map you saved.

Only now replug in your coils/ignitor.

curly

iTrader: (25)

Hmm, listen to Wayne curr or braineack....decisions, desicions. I guess I'll go with braineack (sorry Anton). So I'll do all that without a boot jumper, unless it seems to be not communicating, in which case I'll add the jumper and everything will be hunky dory. Gotcha.

Braineack

iTrader: (62)

I WIN! I may now log off the internet.

wayne_curr

iTrader: (8)

Copy/paste those instructions into the DIY sticky already!

AbeFM

iTrader: (3)

Ask Joe. I remember he found either one set of coils or one set of injectors wasn't firing. Don't know if that's the issue.

curly

iTrader: (25)

Followed Braineack's instructions to the T, except I already had EasyTherm downloaded from the big download packet off diyautotune's website, I also changed the coolant sensor to the more adjusted readings provided in this thread:
https://www.miataturbo.net/forum/t7000/
I re-downloaded the MSPNP no-afm msq, changed the dwell settings and inverted my spark.

Still running on cylinders 1&4.

swapped the COP from 1 and 2, and it's still only running on 1&4. So at the very least I've verified that the COP from cylinder 2 is good, I'm assuming if I swapped 3 for 1 or 4 I'd get the same result. I can unplug the trigger wire for 2&3 and nothing happens, unplugging the trigger wire for 1&4 kills the engine. WTF.

EDIT: originally I changed my settings in the .ini file to say that I was running an AEM_NON_LINEAR wideband. I just checked and that never changed, wouldn't it go back to default if I reburnt it? Or did I only do the .s19 file.

Braineack

iTrader: (62)

its either the circuit of the board itself or the cop harness.

wayne_curr

iTrader: (8)

Thats what i'm thinking. I'm hoping to hell that a solder joint didn't come loose in shipping. It went straight from my running car to the box I shipped it in so I didn't fiddle around with it any more after I tested it.

Braineack

iTrader: (62)

**** like that can happen.

curly

iTrader: (25)

Just tore it apart, and everything seemed fine. Checked all the little wires and such right there for shorts, didn't see any. Pulled a few wires away from each other to see if it was a short I couldn't see, but after just trying it again, no difference. If it would help at all, I could download some pictures I took of the underside, in the LED area. It's my GF's camera, (otherwise they'd be iphone pics) so I'd have to wait a half hour or so till she could tell me where the USB cable is.

wayne_curr

iTrader: (8)

Upon close inspection of those joints, you'd know if something was loose so that must not be it.

I would like to see pics of your wiring harness if thats at all possible. At this point, the only thing I am unsure about is the wiring harness.

Worst case scenario A: You ship it back to me for inspection and another test.

Worst case scenario B: My gf and I take a vacation to Oregon for a weekend =P

curly

iTrader: (25)

On the stock harness, what is the thick, shielded, orange wire? It's coupled with a thick sheilded grey wire.

wayne_curr

iTrader: (8)

What pin does it connect to on the ECU plug?



I dont see any Orange wires on there (on the stock harness anyway).

curly

iTrader: (25)

Pin nothing. It was caught up in the electrical tape that I cut back to expose the wires, and is cut. I didn't just accidently cut it though, it has a female spade connecter on it from who knows when. I'm worried it went to one of my old bandaids (bipes?) then from there back to the ecu harness, and when I unplugged the bipes I never plugged it back into this. But the strange thing is I drove the car 35 miles after I took the 02 clamp and bipes off, with just the BEGi AFPR, so it did run fine.

curly

iTrader: (25)

Does it matter that I'm running straight water with the mazda coolant sensor? It's been reading suspiciously high. Manifold (IAT) will be reading 56*, the coolant will ready 80-90ish, with just a minute or two of idling, which seems oddly high. I've seen a lot of settings in MT about idle control and their settings, wondering if that might have something to do with it, it does rev fairly easily. Anton and I did some more tests tonight over the phone, trying to get a 5v reading off the 2&3 trigger wire with my voltmeter, but I'm pretty sure you need an oscilloscope for that kind of quick on/off voltage. I should repeat those tests at a higher rpm. Wouldn't solve the difference in rpm between the dash and MT, but I wonder if I'm actually getting spark up higher, I just haven't wanted to rev it higher. Could just because it seems smoother once it's not running at 200rpm.

Joe Perez

iTrader: (8)

The physical size of the resistor is related to the wattage rating, not the resistance value. A 1 ohm resistor and a 100,000 ohm resistor will both be the same physical size if they both have the same watt rating.

A 1k 1/4 or 1/8 watt resistor will be fairly small.


The middle LED, by default, indicates that the MS is in warmup more.

The outer LEDs should be OFF most of the time if you're running the new-n-improved spark output circuit and spark out = Inverted. They'll flicker / glow dimly with the engine running, and should be totally dark with the engine off.

Never a bad idea, though the new spark out circuit was designed, in part, to alleviate this requirement. It causes the spark out drivers to be off when the CPU is inactive.




That was in Fuel Table Selection, and he's got it correct.



Shouldn't.


I hate problems like this. They're so hard to find when I can't touch the car...

wayne_curr

iTrader: (8)

Joe, i've done this mod twice and i've gotten both LEDs lit like normal on both squirts. As long as the key is on, the lights are on...running or not.

The only thing I cant wrap my head around at this point is what is different about his car/setup. I tested this megasquirt on my car, worked fine, put it straight into the box it was shipped in and took it to the UPS store.

Joe Perez

iTrader: (8)

Hmm.

(thinks for a while...)



When Spark Out = Normal, then the "resting" state of the CPU pins is high. This causes Q6 / Q8 to turn on, allowing current to pass through D14 / D16 and pulling down the outputs. So a "normal" MS would have the LEDs on when not running.

When we set Spark Out = Inverted, this causes the resting state of the CPU pins to be low. In this state, Q6 / Q8 are off, no current flows through D14 / D16, and a positive voltage is applied to the bases of the two extra transistors, causing them to conduct across the C-E junction, thus pulling the outputs to ground. So in this scenario, those LEDs should be off.


What am I missing here? I ought to know how this damn thing works...



Curly: measure with a voltmeter to see if you have +5 on the two spark outputs (relative to ground) with the power on but the engine not running. If the coils (or igniter) are not connected, you'll probably see about 0.2 to 0.3 volts here, which is the VF (or more precisely, VCE sat) of the transistors.

As you crank the engine, you should see blips of +5 on the outputs. Probably too fast to get a solid reading with a voltmeter, but you should at least see something. Try it with the coils not connected, so you get as accurate a read as possible.

wayne_curr

iTrader: (8)

Your logic is sound, as always Joe. Here is a pic of his spark circuit:


I've looked it over and over again for a mistake and dont see one.

Also, we tried hooking his DMM up while cranking and he barely got any voltage reading at all on either pin but from what I remember the little bit he got seemed consistent between the two spark output pins. Its just too fast for a DMM to get a reading.

Braineack

iTrader: (62)

pretty sure the LEDs stay lit fully regardless of rpm with joes mod with power on.

looks like one of the 330 resistors might be touching the solder joint next to it, off the yellow wire (on the right most).