New build w/ MSnS, Crane 12s etc. - cyl # 1 problems, WHY???

[DaJackal]

Ok, now we have a problem with our project '95 GS. I'll try to explain the situation as well as possible, so pardon me - this is going to be a long story.

First we installed MSnS to the car (MS-I Extra v2.2). We did it according to MB's walkthrough except that we did the IAT & MAP vacuum line fittings in a bit different way (we unplugged the EGR vacuum line since we removed that s**t and put the MAP sensor vac line there instead, the GM IAT sensor went in to the place of the stock IAT with the help of an adapter) and soldered the injector wires directly to the stock connector so we didn't need fuses for them. This walkthrough can be found here: 2GNT.com - MSnS_INSTALL_(parts/wiring_only!)
After some problems with the crank signal (processor had one broken clip, got it to work with a spare processor from a friend) we had it running well enough. Then we parked it in order to do a cam & valvetrain upgrade and a timing job.

So we installed Crower retainers & valve springs, Crane 12s, OBX cam gears, PT lifters and 2.7 rockers for the valvetrain. Then we used the PartsDinosaur mechanical tensioner kit and did the timing job. The OBX cam gears had their timing marks way off so we used the stock gears to get the timing in place and did new marks by ourselves. The timing was half a tooth off (where had been a tooth w/ stockers now there was a "valley") like this so we turned both cam gears 4 degrees counterclockwise in order to get both cams correctly. Parts of the MSnS were not touched during this.

Then it was time to fire up the engine. We started with the MS-I processor and the program that we had used to drive the car to garage. The car would run only with 2 cylinders - 2 and 3. Ok, 1& 4 and 2 & 3 are in the same ignition group as well as in the same injector group, so we thought that there was a problem with the injector wiring (wideband showed that the car is running very lean). My friend checked the MS with a simulator he has. No problems with that.

Next we went throught the injector wiring and did a continuity check. Everything was fine. Then the spark wires, first to the ignitor (fine) and from the ignitor to the coil (fine). Then we tested the coil towers (12-13 kohms both pairs = fine) and then the injectors (12.1-12.4 ohms each one = fine). We took out injector plugs while the engine was running - removing #1 had no effect and removing #4 had some effect. Then we swapped the plug of the injector #1 to go to the injector #2 and left the injector #1 and injector plug #2 unconnected. No effect. Cyl #2 works correctly. ?

Then we tested spark with a timing light. Fine - or at least every cylinder has some spark. Then we pulled a infra-thermometer and started measuring exhaust manifold temperatures. 2 & 3 temps were even, #4 temp was somewhat lower and #1 temp was really low. Then we did some adjusting to cam gears. #4 improved a bit, but cyl #1 still had nothing happening. Well, time to pull the plugs and do some checking.

We pulled the plugs - #1 was wet so the injector works, all the others were dry as they should since they worked. We swapped the plugs between #1 and #2 - no effect. #1 still dead, or at least close. Then we swapped the plug wires between #1 and #4 - no effect. #1 still almost dead.

Time to do a compression test and fear for the worst. Even 170 through all cylinders. SO, what can cause this???


I have a spare coil pack that I get my hands on tomorrow. It is now the suspect #1 though we tested it to be fine according to ohmmeter readings. IF swapping the coil pack won't help, what could we try next? All the connections have been gone through many times. My friend is an electrician and did very well soldered connections everywhere so I trust him in these. The MS and its connector have been checked too. If changing the coil pack won't help I think I'm gonna weep.

Remember: The car ran well with MS when we parked it and started doing the valvetrain & timing job.

 

[fun kart kid]

You might try swaping injectors if number 1 is sticking open that could be the problem. Also did you put new sparkplugs in when you upgraded the cams sometimes you can have a plug bad out of the box.

 

[CODE4]

I assume you are running batch fire for the injectors? Additionally, Injectors 1 & 3 should be paired, and 2 & 4. Not the way you have listed.

Ref: MSnS Extra Ignition Hardware Manual

You must be logged in to view this image or video.

Next, where did these cams come from? Are you 100% sure they are for the Eclipse and not a DOHC Neon? Also, did you verify the static timing using the locating holes between cylinders 1 & 2 facing the 12 o'clock?

 

[DaJackal]

You might try swaping injectors if number 1 is sticking open that could be the problem. Also did you put new sparkplugs in when you upgraded the cams sometimes you can have a plug bad out of the box.

That injector sticking open could be possible. Good point.

We put in new spark plugs of course, but as I stated in my first post we tried swapping cyl #1 and #2 plugs with no effect, so a broken plug can't be the problem.

 

[DaJackal]

I assume you are running batch fire for the injectors?

Batch fire? What is that? I didn't make the MS connections myself and I don't understand its life too much, sorry. Just watched and handed my friend pliers etc.

Additionally, Injectors 1 & 3 should be paired, and 2 & 4. Not the way you have listed.

Sorry, my mistake. This must be the way my friend wired them. I just remembered that the injectors are divided to 2 groups and made the mistake of assuming the groups are the same as with spark.

Next, where did these cams come from? Are you 100% sure they are for the Eclipse and not a DOHC Neon?

Cams came from another tuners member. I confirmed from Crane that they are correct, using the part numbers ground on the cams.

Also, did you verify the static timing using the locating holes between cylinders 1 & 2 facing the 12 o'clock?

You mean the holes in camshafts? Yes we did.

 

[CODE4]

Hmm, it seems you have done pretty well so far for troubleshooting. It sounds like the Cyl 1 & 4 side of the coil pack is suspect, or at perhaps the ground signal is.

Swap in that spare coil pack you have and see how it works. Next I would double check the grounding wire for cyl 1 & 4 - Pin 3 on the coil pack leading to pin 36 on the MegaSquirt connector. Ensure that the solder connections are solid, as a poor connection will create poor grounding signals for that side.

 

[DaJackal]

Swap in that spare coil pack you have and see how it works. Next I would double check the grounding wire for cyl 1 & 4 - Pin 3 on the coil pack leading to pin 36 on the MegaSquirt connector. Ensure that the solder connections are solid, as a poor connection will create poor grounding signals for that side.

Grounding wire? Pin # 36 on the MS connector goes to cyl 2 & 3 ignition (or 1 & 4 ignition, not sure) and is not a grounding wire. (In fact it goes to the 4g63 ignitor unit and from there it goes to the coil.) I checked the grounding wire of the ignitor unit, it was ok too. And I don't believe that the ignitor unit would be broken.

Swapping the coil didn't help at all. Could the coil connector (wiring side) be faulty? I sprayed some connector cleaner to all connectors but I know this won't help.

I guess next we have to take a look at the ignition with an oscilloscope. If that gives nothing useful we will solder the ignition wires back to stock so that the MS only takes care of the fuel injection.

Any help/input will be appreciated!

 

[CODE4]

Grounding wire? Pin # 36 on the MS connector goes to cyl 2 & 3 ignition (or 1 & 4 ignition, not sure) and is not a grounding wire. (In fact it goes to the 4g63 ignitor unit and from there it goes to the coil.) I checked the grounding wire of the ignitor unit, it was ok too. And I don't believe that the ignitor unit would be broken.

The way the ignition coil works is that the low voltage side charges (dwell) and when the field collapses it creates the high voltage/low current to fire across the spark plug. So pin 2 (middle pin) of the 420a coil supplies the +12v for the coil as well as the high voltage side for the spark plugs. Pins 1 & 3 are grounds, and receive ground signals from the ECU (stock or MS, does not matter). So what I am saying is that pin 36 of the MS connector is the ground trigger for Cylinders 1&4 (pin 3 on the coil connector).

Did you mention in your original post that you are using the 4g63 ignitor? That is an important detail not to be left out. Check your wiring between the MS connector and the ignitor, and possibly swap that out to test if it is faulty.

 

[VelocitàPaola]

Where did you get your MegaSquirt ECU from? Can you post a couple of clear pictures of the top and bottom of the circuit board? You said it's a v2.2 board and you mentioned pin 36, but there's no internal connection for pin 36 on v2.2 boards. It's possible to use pin 36 by soldering directly to it, but most individuals use X11, X12, X13, or X14 instead.

 

[DaJackal]

The way the ignition coil works is that the low voltage side charges (dwell) and when the field collapses it creates the high voltage/low current to fire across the spark plug. So pin 2 (middle pin) of the 420a coil supplies the +12v for the coil as well as the high voltage side for the spark plugs. Pins 1 & 3 are grounds, and receive ground signals from the ECU (stock or MS, does not matter). So what I am saying is that pin 36 of the MS connector is the ground trigger for Cylinders 1&4 (pin 3 on the coil connector).

Ahh, I see.

Did you mention in your original post that you are using the 4g63 ignitor? That is an important detail not to be left out. Check your wiring between the MS connector and the ignitor, and possibly swap that out to test if it is faulty.

I referred to the MSnS wiring guide by MB that I linked in my original post. There the wiring is made this way using a 4g63 ignitor. I'm sorry I didn't emphasize this enough in the post itself, I thought it was the standard way to wire it. Continuity checks between this ignitor wiring and MS have been done and passed. We don't have a spare ignitor atm but I don't think the ignitor is the problem.

 

[DaJackal]

Where did you get your MegaSquirt ECU from? Can you post a couple of clear pictures of the top and bottom of the circuit board? You said it's a v2.2 board and you mentioned pin 36, but there's no internal connection for pin 36 on v2.2 boards. It's possible to use pin 36 by soldering directly to it, but most individuals use X11, X12, X13, or X14 instead.

I got it from a member - if I remember his nick correctly it was jesterscorpion. He sold it to me as MS-I Extra v2.2. My friend has the unit atm but I can ask him to take some pics of the circuit board and send them to me.

But please remember: The car ran fine with the MS with exactly the same program before the timing job, cam upgrade & valvetrain work. Timing has been checked 2 times and the markings are in place so the belt hasn't skipped.

Also a timing light shows that cyl #1 and cyl #4 do get at least some spark. This is also backed up by the thing that if we let the car idle we get the following temperatures from the exhaust manifold pipes:

cylinder temp (degrees Celsius)
#1: 60
#2: 120
#3: 120
#4: 100

Now if we rev the engine a bit the temps rise quickly to:

cylinder temp (degrees Celsius)
#1: 100
#2: 170
#3: 170
#4: 155

I think this can't be explained by conduction since the temps rise so quickly to these numbers. So, cyl #1 must have at least something happening.

 

[DaJackal]

Ok, after 2 hours of studying the MS I might have gotten a hold of this.

We did the wirings according to this: http://www.symtechlabs.com/support/docs/420A_MS_Wiring_Guide.gif

Here it shows to connect pins 31 and 36 to coil - to cylinders 1&4 and 2&3 respectively.

And we should have done it like this: http://www.msextra.com/doc/ms1extra/MS_Extra_Hardware_Manual_files/ms2gntwiring.gif

Here this tells us to connect these to pins 25 and 27 (in case of v2.2) or to pins 6 and 36 (in case of v3.0).

My theory:
-The MS unit is assembled in a slightly different way than Symtech does it
-The Symtech version is meant for direct coil drive (ie. no ignitor in between) (?)
-Since the cyls 2 & 3 seem to be working it means that the MS is actually v3.0
-So, in order to solve the problem we only have to change the 1&4 coil wire from MS pin 31 to pin 6

Still this doesn't explain why it would have worked before we tore the valvetrain and stuff, but I guess those pins had some activity and in that way resulted in a some kind of spark, not just strong enough this time? Comments are welcome, please tell me what do you think about my theory.

 

[VelocitàPaola]

-The MS unit is assembled in a slightly different way than Symtech does it

Probably... there's no standard way of doing it.

-The Symtech version is meant for direct coil drive (ie. no ignitor in between) (?)

Correct.

-Since the cyls 2 & 3 seem to be working it means that the MS is actually v3.0

Not necessarily, but the pictures will help us confirm this.

-So, in order to solve the problem we only have to change the 1&4 coil wire from MS pin 31 to pin 6

Again, not necessarily. We really need those pictures...

 

[DaJackal]

My friend has been off town the whole weekend so I haven't gotten a chance to get the MS from him. Going to sleep now, but I'll try to fetch it tomorrow and take the pictures of the board asap in the morning so that in the afternoon we could get into fixing the problem!

 

[DaJackal]

Pictures can be found here: Pictures by DaJackal86 - Photobucket

It is a v2.2 board, but I don't know how it has been assembled. Paul, can you tell me which pins should we use just by looking at these pictures?

 

[VelocitàPaola]

It is a v2.2 board, but I don't know how it has been assembled. Paul, can you tell me which pins should we use just by looking at these pictures?

Yup. The spark outputs are on pins 31 and 36. It looks like the output for cylinders 1&4 is on pin 36, while the output for cylinders 2&3 is on pin 31. There are no internal ignitors, so you have to use the 4G63 ignition module.

The input circuit appears to be correct too.

It's an extremely simple setup, so if the left and right LED's are blinking and there's no spark, there's definitely a problem with either the ignition module or the wiring. If the LED's aren't blinking and you're sure the settings are correct, the processor is damaged.

 

[DaJackal]

Goddamnit. Everything has been checked. We have tried both the current processor and a brand new MS2 daughterboard. Plus the leds are blinking.

Ignition wiring has been checked. It seems like we'll have to recheck though. The ignition module is the only thing we haven't tested. Is there a way to test this module by measuring resistances or something like that?

Also, would it be possible that the coil pack connector plug is bad? Are there other cars that would have the same connector as the '95 model?

 

[VelocitàPaola]

Ignition wiring has been checked. It seems like we'll have to recheck though. The ignition module is the only thing we haven't tested. Is there a way to test this module by measuring resistances or something like that?

Not that I know of, but there's probably a test procedure in the FSM. My money's on a bad ingition module.

Also, would it be possible that the coil pack connector plug is bad? Are there other cars that would have the same connector as the '95 model?

It's possible, but I don't think it's too likely. '95 Neons and Avengers should use the same connector.

 

[DaJackal]

Not that I know of, but there's probably a test procedure in the FSM. My money's on a bad ingition module.

...And a bad ignition module it was. Loaned another one from a friend and tried with it - problem gone. Runs strong on all 4 cyls.

Thanks for everyone, now it's tuning time!

 

[bullettdsm]

Finally! Sweet. Good luck on the tune.

MB