Cleaned up Voltage, still getting secL

[jsdevel]

I've been battling resets since I've installed my MS2/3.57. I've run separate power feeds to the LC-2 wideband controller, the ECU, and the fuel pump relay. I've run 4 gauge wire from my alternator to the starter. I've cleaned all grounds. I'm using resistor type plugs, and I've tied all wires away from the COP wires.

At this point I'm really at a loss and I'm hoping that someone can help. Attached is a log file from my most recent run. You'll notice secL reset to 0 at around 422 and 560.

The only solution I can think of at this point is to run separate ground wires from the battery to the LC-2 and the ECU as they both share the same ground point with my lowz injectors, COP etc.

Any help is appreciated!

[rickb794]

"The only solution I can think of at this point is to run separate ground wires from the battery to the LC-2 and the ECU as they both share the same ground point with my lowz injectors, COP etc."
Where is that ground, I ask?
How is it that your injectors have their own ground?
Have you used a production DIY harness (or equivalent) that has multiple ground wires (connected to pins 14-19)?
You are not running a Relay board, correct?

Have you considered installing P&H board(s)? (separately grounded)

From Section 3.2

"As a check, with the Megasquirt connector unplugged, ensure that the sensor grounds have no continuity to
engine/body ground. Your sensor readings will be junk if they do have continuity - the sensors must ground at
the Megasquirt only."

Also the LC-2 heater ground should not be made at the MS, it should go to the block. No ground connections should be made at the battery!

Another possibility might be to use the mod shown in 13.6 of the v3.0 manual?

[jsdevel]

I am using the P&H board for the lowzs. Would the 13.6 mod still be needed?

I'm not using a relay board.

[rickb794]

So you have a separate ground from the P&H board to the block?

No the 13.6 mod is to do partially what the P&H board does with the MS injector drivers.
It isolates the ground path from the main board.
The 0.1uF cap across the boot loader header would probably help though.
Depends on how the noise is getting into the MS.
What are you seeing in the logs?

[jsdevel]

So you have a separate ground from the P&H board to the block?

No the 13.6 mod is to do partially what the P&H board does with the MS injector drivers.
It isolates the ground path from the main board.
The 0.1uF cap across the boot loader header would probably help though.
Depends on how the noise is getting into the MS.
What are you seeing in the logs?

I have separate grounds coming from the DB15 connector going to the block for the injectors yes.

In the log that I attached, everything seems normal to me. Voltage is steady and then it will randomly reset. I tried bypassing the ignition switch to rule that out, but the same thing is occurring.

Attaching the latest logs from today. You can see around 76 seconds that voltage is flat around 14.1 and then I get a reset. I'm beginning to think it's something internal inside the MS. What would be the main circuits to test? Is it possible I have a corrupted tune?

[jsdevel]

"As a check, with the Megasquirt connector unplugged, ensure that the sensor grounds have no continuity to
engine/body ground. Your sensor readings will be junk if they do have continuity - the sensors must ground at
the Megasquirt only."

I did this and no continuity was present.

[rickb794]

I could not open your log.

Are you getting resets or sync loss?

What is the sync loss reason (code)?

Standard plug wires (no rocket science low resistance wires).
Are the coils mounted to the motor?
To an ungrounded valve cover that works like a broadcast antenna?

[jsdevel]

The first log I attached had to be gzipped. Can you try opening the log in reply #5 herein? That file is in normal format?

I'm getting resets. SecL resets to 0.

After the reset I get a sync loss reason of 17.

I can check the resistance of the plug wires. It should be near 0 right?

LS2 coils are mounted on a powder coated bracket attached to the valve cover that may not be grounded. Could that be an issue?

[rickb794]

Yes the log I tried to open was in post 5

LS2 coils have 4 wires, yes?
Power in
trigger
trigger return
power ground
The last 2 are not interchangeable.

Where does the power ground lead go?
Should be short and to the block.
Are the valve covers actually grounded?


Does the coil power come from the same place as other EFI parts? (should come off the fuel pump relay)(which should be powered from the battery)

[piledriver]

...No, the plug wires should be around 5Kohms.==resistor wires.
Zero ohms means not noise suppression wires, generally, although the spiral core wires seem to work.
Run resistor plugs too.

Technically you only need one resistance in each line for it to work, but reality usually wants both.

[jsdevel]

Checked the wires. My DMM read 346 ohms. I have resistor plugs. Is this resistance too little?

[jsdevel]

Yes the log I tried to open was in post 5

LS2 coils have 4 wires, yes?
Power in
trigger
trigger return
power ground
The last 2 are not interchangeable.

Where does the power ground lead go?
Should be short and to the block.
Are the valve covers actually grounded?


Does the coil power come from the same place as other EFI parts? (should come off the fuel pump relay)(which should be powered from the battery)

I'm reattaching the log to this post. You can see around 76 seconds that voltage is stable, yet secL resets to 0.

LS2 coils have 4 wires. 12V power is coming from the fuel pump relay. Fuel pump relay power comes from the battery through the factory fuse panel. 5V trigger comes from MS (using FETs internally). Trigger ground is connected to sensor ground. 12V ground follows the harness back to a common terminal that the MS shares with the LC-2 controller and injectors. The terminal is grounded to the engine block with an 8 AWG (4' in length). Would this be adding interference back to the MS? Should I re-route grounds to the block as short as possible?

Valve cover may not be grounded. Head is grounded. Valve cover has a lot of black paint that may be preventing it from being grounded.

[slow_hemi6]

It would not be that the resistance was too little, the low resistance just indicates that they are not suppression wires. Suppression wires minimize EMI and RF which is beneficial.

[jsdevel]

It would not be that the resistance was too little, the low resistance just indicates that they are not suppression wires. Suppression wires minimize EMI and RF which is beneficial.

Here is the LS2 bracket/wire setup I currently have in place: https://yoshifab.com/store/8v-gm-ls-coi ... acket.html



Assuming they're not suppression wires, are there any that you would recommend? Would the Zone have a set if I asked for them? If the lack of suppression wires are causing secL to reset to 0, what would I be seeing in the logs?

Per the Setting Up manual, I have made sure to keep the plug wires at least 1" away from the harness.

[rickb794]

If those wires are spiral wound type, most likely not suppression type.

Probably low resistance

This is reasonable "distributed resistance type wires have a resistance value of 3,000 to 12,000 ohms per foot. "

For a short wire like the LS maybe even as low as 1k.

Belden wires from NAPA have always been one of my favorites.

Just avoid the low resistance wires, it is a marketing gimmick.

[piledriver]

Last time I bought wires, I picked up 4 long (11") wires from the local HD dealer, HD brand for twincam, $3 each.
The "short" wire for the Twincam is 3x the cost...

These were carbon impregnated core type.
The boots are a little tight but they work.

On a v8 you might do better with a replacement set from autozone or NAPA.

[jsdevel]

I'll be sure to bring my DMM with me ;). Is there a way to confirm this is the issue before going this route? Would there be a tell tale sign in the logs other than secL resetting to 0?

[jsdevel]

I bought an AM radio and placed it next to the wires. I could hear very faint crackling from the wires. I did notice that my innovative wideband controller was outputting noise like crazy though! I could follow it all the way along the ground cable that the MS is connected to up to where it connects to the block.

On my other car (non MS) with a points distributor the noise from the plug wires was much more audible, like a steady drum roll.

[piledriver]

I bought an AM radio and placed it next to the wires. I could hear very faint crackling from the wires. I did notice that my innovative wideband controller was outputting noise like crazy though! I could follow it all the way along the ground cable that the MS is connected to up to where it connects to the block.

On my other car (non MS) with a points distributor the noise from the plug wires was much more audible, like a steady drum roll.

Kudos!

No one ever tries it, it does help ID the sources pretty well.
The commercially made wideband is probably the last place folks would ever think of as a broadcast source.

Put the antenna down and attach an insulated wire and alligator clip... you can probe "grounds" and 12 v feeds with that.
You will probably learn something.

Also note that getting the power and ground connections perfect (and protected) at the battery matters.
That's your best filter. Vaseline as protectant works as well as the commercial spray on junk for pennies.

[jsdevel]

Based on how much noise the wideband controller is putting out, I think I'm going to mount that away from MS first before moving onto the wires. I can see now why the Innovative manual states that the controller shouldn't share a circuit with any other components. It suggests having a dedicated ground to the battery and a dedicated fused power feed.

I'm thinking the best spot to mount it is near the battery where heat and water isn't much of an issue, but I dunno. I just want this thing away from my speakers and ECU! I now have solid suspicions that this has been the reason for the high pitched hum I hear on the road.

Any suggestions on mount points for the controller are much appreciated.

I will report back. Thanks for the help!