Random Engine Stalling

[activethunder32guy]

OK - update: (Losing my mind.)

I swapped main harnesses a few weeks ago and the problem seemed to go away, so I put a new harness on the engine that was acting up. After a good run in very rough conditions (about 1.5 hours between idle time and running at speed), no issues.

Since then, I've run the boat 2 more times and had the engine stall frequently at an idle - same type of symptom. I haven't had it happen in mid-range for a while, but it happens at 1500-2000 rpm, so I think that rules out IAC issues... If I'm very quick with the throttle at an idle, I can catch the engine and it bounces back almost instantaneously.

I'm basically out of things to try from the wiring standpoint: I'm currently running separate fused jumper wires to:
1. The Holley DIS (this is required by the manual, so nothing odd about that)
2. The Holley computer (again, per the standard installation)
3. The 12V switched key power (this to avoid any question of a key switch issue or anything else from the original boat harnesses)
4. I pulled the Holley relay out and am directly connecting the +12V fused wire to green fuel pump + wire/red fuel injector + pin in the relay socket. So the pump runs all the time when the fuse is plugged in.
5. I've replaced the ignition harness, swapped crank triggers (made sure again it was set exactly per instructions).

I have regrounded the entire engine, new wires, everything is wired exactly as specified.

The datalogs are showing basically nothing - and that doesn't make sense - possibly indicating a root cause. I'm not 100% sure, but the log keeps dropping out with no sign of rpm drop... like the computer is resetting before the rpm cuts out. I would run the engine for 30 seconds and have a 5-8 second datalog file. My thoughts are could I have a bad main power connector (the connection at the ECM may be faulty?) Otherwise, what could reset the computer? Is it possible for the computer to reset and get it's bearings fast enough that it could sometimes keep running with a quick jab of the throttle? The boat will sometimes idle for a whole trip without faulting, but more often it will error within 2 seconds of starting... the engine will start, jump to the target rpm, then try to cutout, then either bounce back, or kill. Basically if it cuts at 1500-2000 rpm, it will not kill, just stumble to 700 rpm then go back to 1000 within a second or so. Any chance a TPS sensor could cause this?

Ever since I changed my ignition dwell, which made a lot of things better, I haven't been able to see a system log with a missed crank trigger, and I don't think I've had an issue above 2500-3000 either, but I haven't run there very long either. When the missed triggers were happening, the computer would log right through it, now it seems I may have a different issue with similar symptoms (very possible it's always been there).

Another question - is it possible my issue is purely software based? I did swap ECUs, both have the latest firmware that matches the latest software on the computer. I've been working both engines with completely separate Global Folders & ICFs...as I should. However, could I have a corrupt Global Folder? I could always take the good engines' folder and run the bad engine off that base global folder/ICF set.

[Danny Cabral]

Basically if it cuts at 1500-2000 rpm, it will not kill, just stumble to 700 rpm then go back to 1000 within a second or so.

I have another thought. Is this a wideband O2 sensor application (Closed Loop)?
If so, try running the engine in Open Loop mode and see if it still does it.

[activethunder32guy]

Yes - good idea. Still running closed loop. I don't feel very confident running the whole rpm range without closed loop yet, but I certainly could try starting closed loop above 2500 rpm.

Generally speaking, do you think this could cause a reset in the computer? Or are you thinking the sensor could be getting a bad read, causing a bad rapid adjustment?

In that case, I could also try running a lower percentage for closed loop at idle, just not sure how low would smooth out transients.

[Danny Cabral]

Or are you thinking the sensor could be getting a bad read, causing a bad rapid adjustment?

Yes, I'm thinking about the WBO2 sensor spiking full lean or rich.
Where is the WBO2 sensor located, and what's the Advanced Control set to?
Also, have you tried swapping the WBO2 sensor from engine to engine?

In that case, I could also try running a lower percentage for closed loop at idle, just not sure how low would smooth out transients...

5% should suffice, but since we're trying to rule things out, run it in Open Loop.
(Disable Closed Loop Parameters in System Parameters).

[activethunder32guy]

First to answer your questions: The WBO2 is about 18" back of the collector of the header. It is a marine header, but is water jacketed and does not mix water with exhaust until the end of the tip. I have run Bosch WBO2 sensors for several years to help with carburetor tuning with good success. Advance control is set at 3. I did try swapping 02 sensors engine for engine with no change.

I tested the engine on the trailer tonight... made several more diagnostic changes. I'm about out of ideas... Nothing I did has changed anything.

The good news, the only good news, is that the engine seems to be stalling within the first 10-15 seconds of running today. A few times it ran for a few minutes, it was able to warm up to operating temperature so that I could test with and without the close loop compensation. As mentioned before, it has run for hours straight without faltering too. When it's running, it runs perfectly, all cylinders, clean AFR - smooth idle.

Changes made:
1. Swapped computers checking for a corrupt file/computer issues - also wanted to rule out me messing up a setting: moved the computer from the good engine to the bad one, I left the program, just made adjustments for the DIS (I'm still running MSD on the good engine), made sure the ignition settings, timing sync, and re-calibrated the TPS. No change.
2. I tried removing the closed loop o2 from the program. Started it up, fired right up, ran for maybe 30 seconds to 1 minute, then instantly shut down. I was watching the dashboard, AFR was reading 13-14 the whole time, good fuel pressure and all of the other sensors were good.
3. Installed another crank trigger, made sure the gap and location was correct. (I had an extra from the DIS kit)... No signs of damage on the old one (Holley tech had asked if it was rubbing).

This lead me to check the datalogs... here's the thing, I'm not seeing ANY rpm loss/stalls. According to each datalog/SL, the engine runs perfectly, it "stops recording" before an issue occurs. This tells me the computer is shutting off/resetting, causing the engine to shut off. I repeated the datalog checks throughout my trial and error process, but never actually caught the engine stalling. I also tried SL logs, watched the crank trigger, and it was working perfectly... never missed.

First question: What would cause the ECM to stop datalogging and at the same time, stall an engine...my thought is it's losing power and resetting. I watched the logs carefully, no sensor, crank trigger, WBO2, fuel pressure, etc... showed any sign of issue right before the program stopped. I am using 100 FPS for resolution to be sure I'm not missing anything.

-I checked the main power harness connections to the battery, they are textbook, clean and tight.
-I made sure the ECM connector contacts were good and tight on the harness. I also checked the fuse, it was tight.
-As a 3rd test, I removed the main harness from the good engine ECM, which is connected to another battery, and connected it to the bad ECM. No change.
-The 12V switched power is currently directly connected to the battery with a fused cable. I added this a week ago to rule out issues in the boat wiring/key switch. I also have run a multimeter on this during an engine stall, no loss of voltage. Also keep in mind this is my second main harness, so I redid the red/white wire connection to the switched source (and in this case, direct connection to the battery).

In other words, the ECM is getting consistent voltage (14.3V when the engine is running, 12.5V when starting without the alternator), but it's completely missing the events where the engine cuts out... I need to understand what can cause this.

I was thinking about swapping throttle bodies next, to try to rule out the TPS sensor causing an error or the IAC for some reason. But with the datalog showing nothing, when the engine is killing, I really need to go after that first.

[Danny Cabral]

First question: What would cause the ECM to stop datalogging and at the same time, stall an engine...my thought is it's losing power and resetting.

I don't usually suspect this, but at this point, it does seem like that ECU is being affected by RFI/EMI noise.
Your situation reminds me of this thread below, where the EFI main harness required shielding modification due to interference.
I've only heard of this requirement twice, but I still think it was due to high voltage electrical components too close to the ECU or harness.

Modified Holley EFI main harness discussion in this thread:
http://www.yellowbullet.com/forum/sh...57#post9470857 (Read posts by "RAMAIR" - Bob Bales & Joe M.)

[activethunder32guy]

OK, I was afraid of this...
Having read through the forum, I see that it sounds like similar issues... Other than the basic theme of separating wires by type and possibly adding shielding, I'm not totally clear on what would need to happen.
I don't know if I really want to send my harness out to someone that would void my harness warranty, possibly use up the rest of the summer, when I have intimate knowledge of every wire in it myself at this point. I'm willing to modify the harness anyway necessary to get it appropriately shielded...

I purposely set up the locations of my coils to be away from everything. Same goes for the spark plug wiring... The main power wires purposely run low on the engine quite a distance from the higher power wires. The ECM is up high, by itself... I do have exhaust headers that rise up pretty high over the ECM, so maybe the spark plug wires are radiating into the header pipes, causing all kinds of issues?

Keeping in mind that I've bypassed all of the Holley main harness power wires (with wires that very effectively avoid running parallel or even getting close to the main/power harnesses), what key wires do you think I should be rerouting?

More than that, what wires would I want to shield that could actually cause a computer reset?

I assume spark plug wires are the most troublesome component
followed by battery cables/alternator wires...

Some comments about shielding spark plug wires came up... are you aware of any shielding methods/product specifically for a spark plug wire?

[Danny Cabral]

I don't know if I really want to send my harness out to someone that would void my harness warranty, possibly use up the rest of the summer, when I have intimate knowledge of every wire in it myself at this point. I'm willing to modify the harness anyway necessary to get it appropriately shielded...

I completely agree. In no way was I implying you send your EFI main harness to anyone for modification.

More than that, what wires would I want to shield that could actually cause a computer reset?

A couple users have found that opening the Holley EFI main harness to shield & separate the
ignition and sensor wires (from all other wires) has solved previous electrical interference problems.
I'm still not convinced this is necessary. I think this just rectified a component installation error.

I assume spark plug wires are the most troublesome component, followed by battery cables/alternator wires...

I suggest moving an AM transistor radio over the running engine to possibly locate troublesome interference (Link).
I highly suggest you watch this entire video: https://www.youtube.com/watch?v=zIi2_lCcazs (EFI Electrical Noise)

Some comments about shielding spark plug wires came up... are you aware of any shielding methods/product specifically for a spark plug wire?

No, but I'm aware that certain spark plug wires (such as FireCore50), transmit far less electrical interference than others.
The wires should be spaced away from each other and from exhaust heat. Also, are you using resistor type spark plugs?

[activethunder32guy]

Thanks for the video link, that may help a lot... I will definitely dig into that and try the transistor radio tip too.

I am running NGK R5671A-8 plugs. After digging a bit, these are non-resistor plugs...hmm.

There are sections where the wires are within 1" of each other running parallel around to get to one of the banks...
the other bank is about as separate as possible. I am running MSD's 8.5mm Super Conductor wires right now.

[Danny Cabral]

I am running NGK R5671A-8 plugs. After digging a bit, these are non-resistor plugs...hmm.

You should use resistor type spark plugs with an EFI system, to suppress electrical noise/interference.
(Unless it's a race only application and the non-resistor plug is the only heat range & dimensions available.)
Also, certain spark plug wires (such as FireCore50), transmit far less electrical interference than other
brands (low resistance and impregnated with ferrite molecular compound used to absorb RFI/EMI).
http://www.yellowbullet.com/forum/sh....php?t=1325290 (Additional Information - Doug F., Chief EFI Engineer @ Holley)
NGK also offers spark plug resistor caps to use with non-resistor type spark plugs:
http://www.ngksparkplugs.co.za/produ...resistor-caps/ (NGK Spark Plug Resistor Caps)
http://www.ngkpartfinder.co.uk/files...lug_Covers.pdf (NGK Spark Plug Resistor Caps)

There are sections where the wires are within 1" of each other running parallel around to get to one of the banks...

One inch spacing is fine...mine are less than that.

I am running MSD's 8.5mm Super Conductor wires right now.

I use the same spark plug wires without any issues.
Conversely, not everyone has claimed the same experience:
http://forums.holley.com/showthread....-stops-mid-run (Supposedly, an MSD Super Conductor Wire problem.)

[activethunder32guy]

Update:

1. The spark plugs definitely were causing many of the idle stalls - after changing them, the reliability got much better at idle. I still had issues with mid-range rpm. As mentioned before, one engine ran fine with the plugs (NGK R5671A-8). In my particular case, changing to Holley DIS ignition, I was going to use a iridium NGK IX plugs either way, so once I installed the new plugs, the idle portion of the problem was gone.

2. The rest of the issues were with the key switch +12V power. I had been bypassing this wire to rule out multiple variables before the spark plug issue came to light. Changing this back to normal after the spark plug issue was resolved still yielded the mid-range issue. I went back to bypassing the key switch and all problems were solved. The boat has run off this same +12V ignition wire for years before EFI, so I must have a VERY intermittent issue that's loosing power just long enough for the ECU to reset. I will try a new key switch first - all connections have been checked and rechecked after that point. If the key switch doesn't solve the issue, then I will need to look at routing the wire differently - possibly an EMI issue. Either way, I know exactly what wire is giving me an issue, and I can control it.

Thank you Danny for all your help!

I am very pleased with how the engines are performing now. Although my engines were "well tuned" carbureted, the HP multi-port EFI system has made significant differences at an idle dealing with the hot/cold engine with regard to air/fuel mixture. I used to shoot for the middle in the idle mixture. When heading out the channel for 30 minutes, the engine would run a lean AFR (causing some surging and cold-blooded behavior). After running the boat on the lake, the boat would run a rich AFR returning down the channel (fouling plugs). This rich mixture would wash down the cylinder walls with extra fuel, causing premature wear on the piston rings. I can tell already my oil is MUCH cleaner after several good weeks of boating. Mid range and top end performance is excellent, there is no drop in performance whatsoever throughout. Very pleased. I just need a longer summer.

[Danny Cabral]

1. The spark plugs definitely were causing many of the idle stalls - after changing them, the reliability got much better at idle.

Yep, I mentioned this in post #53.

2. The rest of the issues were with the key switch +12V power.

Yep, I mentioned this in post #3.

Thank you Danny for all your help!

You're welcome. I'm glad the engines are running well now.

[RC314]

I have been reading this for a year or so, and just now registered. I'm having shutdown issues as well. Tomorrow I will test voltage at the fuel pump while driving. So one of three things are wrong - bad Bosch 044 pump, relay or HP EFI telling fuel to shutdown.