MS3X Misfire issue

[PeteJ]

I'm running a 418" small block on E85 and am having a consistent misfire issue at all RPM. If I put a timing light on any plug I can see the light go blank on regular rhythm. I'm having a few other "bugs" and am not sure if they are related:

• TPS goes negative once the engine fires

• The engine runs slightly rich no matter what I have in for the target AFR (about .1 lambda too rich)

Attached are the tunes and log files I pulled down.

[PeteJ]

Another few data points.

I'm running NGK UR5 plugs gapped at .035.
-I'm going to try to set the gap at .025 and see if that fixes the issue.
The coils are GM D581 coils.
-I had the dwell set at 4ms and tried bringing it down with no luck.
The engine runs VERY cool at idle.
-Coolant temps never really gets much above 120* and the oil temp gets to about 150*

[Matt Cramer]

TPS going negative and the AFR offset are often grounding issues; where are the TPS and wideband grounded?

What sort of coils and ignition modules are you using?

[PeteJ]

TPS going negative and the AFR offset are often grounding issues; where are the TPS and wideband grounded?

What sort of coils and ignition modules are you using?

The TPS is grounded to the Megasquirt directly. All Megasquirt grounds go the back of the cylinder head with a 2 GA wire running from there to a ground block. The 02 sensor is grounded to the ground block.

I will check to make sure that there is no voltage difference between the ground locations and verify the TPS ground is solidly connected.

I'm using factory GM coils (D581 variety). The coils are driven directly off of the MS3X outputs. I will verify the coil grounding to make sure it is a good connection to the engine block as well.

[PeteJ]

I updated the grounds and made sure everything is grounded properly. I've added up some updated logs, as you can see when I rev the engine the rpm drops to zero every once in a while. Any thoughts on what can cause this, hardware issue etc?

I'm about out of options, I'm about at the point of putting in a conventional distributor with an MSD and call it a day.

[PeteJ]

Looking at the tooth log csv file it appears that there are missing pulses. I brought the file into excel and started a count from 1-36 assuming that the "long pulse" would be the missing tooth and quickly noticed that the "long pulse" was not crossing at a "36" . Am I interpreting this correctly? If so what would cause this, noise? bad VR sensor? anything?

[Matt Cramer]

Can you get a tooth log instead of a composite log? These are a bit easier to use with 36-1 wheels.

[PeteJ]

Attached are some updated logs. I isolated the wires and it appears to be better but it's still there. I can't see the reset in the logs but the timing light skips every once in a while and you can hear the misfire. I'm making a new pickup wire for the Hall Sensor using shielded twisted pair, hopefully that will help out a bit more.

Time is running short for me to make it out to the track this year, hopefully I can get this sorted out soon.

Thanks in advance.

[Matt Cramer]

Unfortunately, neither log seems to have captured the lost sync issue. Try making them longer and make sure it happens when the log is running.

[PeteJ]

I ran a new wire and the reset is clearly visible when I crack the throttle The new wire is routed away from everything else and is grounded at the MS3. You can see it in the tach as well.

[PeteJ]

Looking at this I'm wondering if the board is set up wrong. I'm using a diyautotune 36-1 wheel and threaded hall sensor.

http://msextra.com/doc/ms3/ignition.html#vrv3pull

Am I missing a resistor?

[Matt Cramer]

It appears to be missing part of the Hall / opto circuit, but not the VR conditioner.

The tooth logger showed a bit of a rough start, but didn't capture any problems up to the highest RPM it revved to. I may need a longer tooth log. I'm hoping to find a point where it does something like this:



Or any other point where it fails to fit the "34 short, 1 double height" pattern.

The data log does show a real loss of sync at around 4000 RPM, but the tooth log will give more info on WHY it occurred. All I can tell from the data log is that it's an RPM input issue.

[PeteJ]

Thanks Matt, for now I don't have a priming pulse or enrichment configured for start-up so yes, it does start a rough. I'll try to capture a longer tooth log today and post it. When I have a timing light on it I can see it jump all over when the miss happens.

Would the hall/opto circuit cause the problem and would you recommend starting there?

[Matt Cramer]

If you have TachSelect wired to VRIN - which looks to be the case from that picture - the Hall / Optical circuit is not in play.

[PeteJ]

Thanks Matt, I looked and do have the TachSelect wired to VRin and have a wire from VrOUT to TSEL. When I first configured this over a year ago I was planning on runnign a VR sensor but decided on your hall sensor and wheel instead. I made the appropriate Software changes but failed to do anything on the board. I'm trying to understand what the best configuration is for the V3.0 board with a MS3X.

From the link I provided it looks like the following is how it should be configured...and I'm missing step "c":

V3.0 board - VR Input with pullup for hall sensors, LS2/58X, optical sensors or points
a) Solder a link between VRIN and TACHSELECT
b) Solder a wire between VrOUT and TSEL
c) Install a 1k resistor (any value 470R - 2k2 is likely ok) in the proto area. Connect one end to the 5V hole and join the other end to VRIN with a jumper wire.
d) With a small screwdriver, turn the pots, R52 and R56, about 12 turns anticlockwise (sometimes you may feel a "click" when the end position is reached, they can't be damaged by turning too far.) and then turn R56 back about 6 turns clockwise.

[Matt Cramer]

Our sensor doesn't need a resistor, but you may need to adjust the trim pots a little.

[PeteJ]

Here are some longer log files. In the tooth log when I open it up in Megalogviewer on page 49 I can see a problem with the missing tooth showing up almost twice in a row.

[Matt Cramer]

That's more like it. You have a case of the sensor failing to detect teeth that are there, rather than seeing things that aren't there. Try checking to be sure the bracket is sturdy enough not to have vibration issues, and try a tighter sensor gap.

[PeteJ]

THAT DID IT!!! Thanks Matt.

I tightened up the gap to .005 or so and that took care of the problem. Thanks Matt. Throttle response is crisp, now with a quick crack of the throttle it jumps to over 6000 RPM!

All things willing I'm off to the drag strip on Sunday.

[Matt Cramer]

You're welcome! Thanks for sharing the final result; that can help others who have this issue.