EDIS SAW problem

[Beau M]

Hello everyone. I'm running a 3.57 MS2 Extra with the latest firmware 3. something I think, and a relay board. I am using the EDIS4 module. I can get the car started, it fires right up and the timing is at 10 degrees like its in limp home mode. The squirt thinks it is commanding timing, but no change at the crank with a light. It is reading RPM no problem. It was reccomended I set it up like this:

feed the PIP signal in through the VR conditioner: JP1 in the 1-2 position, J1 in the 3-4 position, set ignition input capture to falling edge in the software

I have PIP going in the Tach port and SAW comming out the S5 port on the relay board.

I believe I have a SAW problem. It could be a hardware issue (loose wire or something) which I would check, but I was hoping you could give me some hints on where to start troubleshooting. According to this link, for output (I'm assuming input is working fine if I get RPM signal)

http://www.msextra.com/doc/ms2extra/MS2-Extra_EDIS.htm

I need to run a jumper wire from J10 to IGN. Any way around this? I would rather not have to solder on my 3.57 board. If I have to, I might as well go direct coil control. Is there a setting in tuner studio that would affect SAW output? I will post a datalog when I get home if that will help.

Thanks!

-Beau

[Matt Cramer]

Normally we would build a V3.57 board with the JS10 to IGN (actually the center hole of Q16 on a V3.57) already there. What do you have set up for ignition output now?

[Beau M]

Not sure, here are some pics while I do some more searching:


http://img.photobucket.com/albums/v247/ ... board4.jpg

http://img.photobucket.com/albums/v247/ ... board3.jpg

http://img.photobucket.com/albums/v247/ ... board2.jpg

http://img.photobucket.com/albums/v247/ ... board1.jpg


Thanks guys!

[SymTech Laboratories]

The jumper from JS10 to IGN is there, so assuming the EDIS module is wired correctly, you probably have a configuration problem. Could you post a copy of your MSQ file?

[Beau M]

Hey SymTech, I just got this thing back from you guys. You said you didn't do anything and it checked out fine, but when I plugged it back in the car fired right up! So thanks for that, the injectors are working great now.

Here is my MSQ file, but I bet it is some sort of wiring problem. Hopefully I can dig into that tomorrow. Thanks for the help so far!

-Beau

[Beau M]

I just checked my wiring, and I have continuity from the MS end of my relay cable on pin 36 all the way to pin 3 at the EDIS module, through the relay board.

[Beau M]

It has been pointed out on another forum that I do not have a 5v pullup installed for my SAW signal. Is this required with my setup? I just found this:

http://www.megamanual.com/ms2/Ignition.htm

Specifically:

The ignition signal going to MegaSquirt-II to the DB37 pin 24, as usual. This is connected to pin 14 of the 40 pin socket (IRQ), and pin #1 of the MC9S12C32 processor.
The signal from MegaSquirt-II™ to the module comes out on:
Pin #17 of the 40 pin socket (labeled IGN), and is connected to the 5th hole of the JP1 header (pin 8 of the MC9S12C32). It MAY be connected to DB37 pin #30 with V2.2 main boards, if jumpered as recommended.

Pin #36 on the DB37 (JS10 jumpered to IGN) for V3.0 main boards if not using the high current ignition driver circuit (HEI/EDIS/etc.),

Pin #36 of the DB37 for V3.0 main boards (IGBTOUT jumpered to IGN) for V3.0 main boards if using the IGBT high current ignition driver circuit.
Note: If you are not using the high-current circuit to directly control a coil, you likely need to to add a pull-up resistor (which allows the signal to go both high and low, rather than floating and low).



You can do this by:

Soldering a 1K Ohm resistor to the hole marked S12 immediately below R57 (this gives a 12 Volt pullup), or to a 5 Volt source on the board for a 5 Volt pullup,
Connecting a jumper wire from the IGBTout jumper to the 1K resistor at S12 hole (at the other end than what's soldered to the board),
Don't remove the IBGT to IGN jumper, just solder the new wire to it.


-Beau

[SymTech Laboratories]

You do not need a pull-up resistor to generate the EDIS SAW signal using JS10. Always reference the MS/Extra manual (not the B&G MegaManual) if you're using the MS/Extra firmware.

http://www.msextra.com/doc/ms2extra/MS2 ... tm#edisint

[Beau M]

Thank you, they look so simmilar I get confused. I was kind of hoping that was the issue, right now I'm at a dead end.

[Beau M]

Somewhat of a result! I hooked everything back up, and am getting movement of the timing marks at the crank. Except, not wont won't idle very good at all, and when I give it a little throttle the timing marks jump all around (even with it set to fixed timing in MS) and it dies.

But, its doing something different, and the SAW signal appears to be making it to the edis module.

Edit. I spoke too soon, for some reason its back to how it was. All I did was remove the SAW wire from the relay board to test my theory and go back to the edis module commanding timing, which it did. Re-connected the wire... and nothing. Still stuck at 10 degrees.

[Beau M]

This is getting a little confusing. I couldn't get it to come out of limp home mode again, where I had a rock solid 10 degrees. Someone on another board suggested my VR sensor polarity might be backwards. I thought if that was the case, I wouldn't get any rpm signal. So, I cut the shielded cable at the sensor and put in a couple bullet plugs so I can swap back and fourth. When I swapped the polarity, no start. Didn't even register RPM in TS. So I swapped them back... and when it started, MS decided to start commanding timing again. I could only keep it running by hand, and the timing was jumping all over the place. I caught a glimpse of the tooth logger, and it looks all messed up. Does this mean I have a bunch of noise on the VR and / or PIP/SAW lines? How do I get rid of that, I'm already using shielded 2 core wire.

Here is the best datalog I could get, its a csv file for some reason.

Halp!

[Beau M]

Well, I'm going to re-wire my ignition signal wires tonight, even if its just so I feel like I'm doing something. Right now I have the shield for both the PIP/SAW and VR leads terminated at the EDIS module to the pin 7 wire, just like every diagram says to do. However, I have been hearing it would be better to ground the PIP/SAW shield where the MS is grounded. What about the VR lead? Should I leave it connected to the EDIS pin 7, ground near the edis module, or connect it to the shielding of the PIP/SAW wires?

I'm also going to pick up one of those in-line noise supressors for the power into the squirt.

I'm getting worried, I'm supposed to be at a car show with this thing in a month.

[Beau M]

I made some... discoveries just now. I totally re-wired my VR sensor and pip/saw lines. Instead of grounding the shielding to pin 7 (at the EDIS), I tied the shielding together at the EDIS module, and grounded the other end at the squirt. (Leaving pin 7 unhooked) This had an effect.. When I switch from fixed timing to use table in tunerstudio, it does what its told and tries to command timing.

But, now the RPM gauge goes CRAZY when that happens, and the car dies. Also, I started getting resets and even a error that there was a problem with the serial cable. This has never happened before. On top of all that, when I went into the car to turn it off, I noticed a loud pop pop pop coming from the speakers. Ugh, I HATE electrical noise.

Datalogs attached

[Beau M]

Well, at least my problem seems consistent now. With it set to fixed timing, it will idle at 10 degrees all day long. As soon as I switch to use table... rpm gauge goes nuts, megasquit resets, and it dies.

I ran an 8awg wire from the engine block to where to squirt is grounded. No change.

I have been reading that I need "resistor spark plugs" and $100 Magnecor wires. Sound correct? I might as well go with LS1 coils if that's the case.

[rickb794]

Have you reloaded the firmware?

Start from scratch with a new download....

I would also try a new MSQ.

[Beau M]

Thanks for the reply! I will try that, firmware re-load seems to be the first step to any fix.

But... New development. Because I sent the squirt out, I wan't sure if the trim pots had been messed with, so I cranked them all the way counter clockwise like you're supposed to with EDIS. I also stopped by the auto parts place and picked up one of those electrical noise filters and put it in line with the power line to the squirt.

I think it did something. I set it for fixed timing, it fired right up and was solid at 10. Then, I selected use table... it tried to die, then even out and was idling! I'ts still freaking out with the RPM jumping, but it was not causing re-sets and I could even rev it a bit, but the timing marks would jump ALL over the place. But it idled an I haven had one reset.

Here is the weird one though. If I go back to fixed timing, it dies immediately, and I can't re-start it. No reset or com issue, just dies. If manually reset the squirt, it fires right up. datalogs and MSQ included.

[Beau M]

I tried re-flashing the firmware, and you know what, I think it helped. It just seems a little smoother when using the table and not fixed timing, it actually idles. But, when you give it some throttle, the timing still freaks out, I even got it to reset again.

I did not create a new MSQ file, I will do that tomorrow, and re-flash again.

I have run into a new problem, and I'm sure part of it my inexperience with datalogging. So, you click Start to, well, stop datalogging right. Now, do you click Stop, then check the box to save the file? Or do you check the box to save the file while its running?

My issue now is when I try and save a log, it looks normal, box pops up, and I click save, and its pointing at the same folder where my other logs are. But when I go to find them, they arent there. Could it have anything to do with not creating a new MSQ?

[SymTech Laboratories]

We recommend switching to "Use Table" for the fixed advance, but more importantly, changing spark output to "Going High (inverted)".

http://www.msextra.com/doc/ms2extra/MS2 ... m#Edissoft

[Beau M]

Thanks, I saw that. I also just watched a tutorial on datalogging, its not exactly intuitive. Anyway, check these out. I didn't change anything, went out to use my new datalogging skills.. and it seemed to run much better! Timing wasn't smooth, but smoother than it was. Thanks everyone for your help so far, it's the community that makes this thing manageable.

[jsmcortina]

We recommend switching to "Use Table" for the fixed advance,

Why? The fixed advance setting was added for just that purpose so you can leave your table alone.

James