Alternating VS Simultaneous inj. causing sync loss??

[Kyle]

I have been chasing the cause of an intermittent sync loss problem with no success. By intermittent I mean it I get a sync loss about every 3 to 5 minutes of running. Idle or driving doesn't seem to make much difference.

I know that when I was still running on the carb with MS only taking data I had no sync loss. Then I installed a GM throttle body (2 injectors) and got the car running, then at some point after that I started getting sync loss. Not sure when exactly, but I made changes to a few things so I have been trying to back track what I might have done.

I recently I changed back to 6 injections per cycle with alternating injections and the sync loss went away (at least it went down significantly. Haven't run it enough to say for certain its gone). I had been running 2 injections per cycle with simultaneous injections. I moved away from the 6/cycle originally because it results in an injector pulse width just below 2ms when at idle, and I was afraid I might get myself into problems later.

I'm wondering if anybody here has any suggestions as to what might be causing the sync loss when switching from alternating/simultaneous injections?

In the process of trying to find the cause of the sync loss I messed with the wiring a little. I pulled the crank position sensor wiring (signal and power) out of the loom and routed it across the fender and around to the sensor so that it is 6 inches or more away from any other wiring in the car. The injector power supply and signal wires are still in the loom though, and so is the power and ground supply for the MS3. Maybe the injector power is creating some feedback in the MS ground or power? I have 10 gauge power and ground, could it be a restriction when injectors are firing simultaneously? Basically everything else goes through that loom for a couple feet on the firewall behind the engine, including the bundle from the MS3 to the DIY relay board, maybe I should try splitting the loom up a little?

I keep coming back to the injector wiring because the injectors, along with TPS and idle valve are the only new wires I have installed since I had clean signal.

Attached a MSQ from simultaneous injections and a sync error log.

Another thing that may be of note, sync error pretty consistently occurs at the 33/34 tooth (36-1 trigger). Not sure what that might mean. Checked the sensor spacing and it is same as the other teeth.

I'm going to go drive it around more and see if I can get it to throw a sync loss with alternating.

Thanks for any suggestions
Kyle.

[Kyle]

update,

I ran the jeep a couple different times this weekend and did get one sync loss event. Log file is attached. So, I'm figuring that the sync loss was due to some signal interference from the two injector power wires being routed right next to the power supply and ground for the MS (zip tied together). Since both injectors are not both firing at the same time when set to alternate, the interference is being reduced and it doesn't have the same effect. Can anybody confirm this makes sense?

I'm going to pull apart the loom and replace with two or three bundles running across the firewall. I'm thinking one for the MS power supply/ground and the low/ no current signal wires, one for the higher current wires (fuel pump, injectors, maybe O2 sensor power supply), and one for the existing engine wiring (fuse box power, headlights, ect). I'll give that a try and then see if I can go back to a 2 injections per cycle - simultaneous setup.

Nobody has told me I'm crazy yet so I'll just keep wandering along.

Any suggestion as to how far apart wires need to be to prevent interference?

[billr]

Honestly, I don't think cross-talk there is your problem. I have my wiring all jumbled together, violating just about every wiring rule promoted here, and have no sync-loss issues, running a 60-2 wheel and no series resistor. I just turned the two pots fully CCW for my VR crank sensor and was done with it. No effort has been made to keep wiring away from ignition coils or plug wires. In fact, the (Hall) cam sensor signal goes through the inside of the coil-drive module, with the sensor power being tied in with the coil power. I am not advocating sloppy wiring, and not "proud of it", but my experience is that the V3.0/MS3/MS3X combo is reasonably noise-tolerant. I do have both my crank and cam sensors in properly shielded cables, though...

[Kyle]

Billr, thanks for the input. I pulled the main loom apart a little more but there was no noticeable effect. I will abandon that route for now.

Also, the crank position sensor is wired in using one of DIYautotune's shielded cables that they sell specifically for that purpose. I looked over the wiring for that again and compared to the info here. I had the shield grounded directly to the engine block at the point the sensor is attached to. Since the info seemed to indicate I should be grounding it at the MS3 I moved it to the same point the sensor ground is connected to (TPS ground on the DIYautotune relay board). This resulted in a significant increase in the sync loss number (every 10 to 15 seconds), so I moved it back to the engine block. I'm not sure why is would cause problems when grounded at the relay board because it seems like that is what the common practice is.

I ran it twice for 15 minutes. Once with simultaneous injections and once with alternating. With alternating I did not record a sync loss in the 15 minutes. With simultaneous I recorded six in 15 minutes. I don't believe that this a problem with the MS cpu. Since it seems like sync loss problems are usually a grounding issue, I took a couple pictures of my wiring. Maybe somebody will notice something I'm doing wrong. Sorry about the poor quality.

This is a shot of the sensor mounting. You can (mostly) see the sensor is attached to a mounting plate I welded to the fuel pump cover plate. It's the white blob forward and below the valve cover. I temporarily ran the wire forward to the radiator and around on the fender well to convince myself it was not getting interference.

image201407150007.jpg

The is a shot of the relay board. It is connected to the MS3 in dash with a DIY patch cable. Power and ground come from the fuse block in the next picture.

image201407150004.jpg

Fuse block is the clear plastic cover on the side of the battery. It is powered and grounded directly from the battery. The primary ground from the battery is directly to the block.

image201407150002.jpg

Anybody see anything that makes them wince?

Since the problem seems to be related to the injectors, could they be the cause of some signal noise if they were damaged? They are spraying fuel, but maybe there could still be something wrong with them that is causing feedback?

Thanks for any input.

[billr]

Post two run logs (.msl), one running simultaneous and one running alternating (no other changes in the .msq).

Also post a tooth log, for both configs.

[Kyle]

I started it up and let it run until the engine was up to temp and then shut it down and checked the injector setup.

Attached are log of the startup with approximately 2 minutes of run time, and a short tooth log. Injector setup is 2 inj. per cycle and simultaneous.

Then I shut it down and changed the injector setup (only the number of injections per cycle and alternating/simultaneous).

Then started again, logging startup and a short tooth log. Setup is 6 inj. per cycle and alternating.

Then shut down and changed again, restarted, and took logs. Setup was 3 inj. per cycle and simultaneous. I'm attaching the last set only because I caught the most sync errors in the last run. It does not include the startup though.

Can only upload 3 files at a time so the rest will follow.

[Kyle]

More logs.

Thanks for the help.

[Kyle]

I may have found a significant contributor to the sync loss problem.

I have been messing around with my grounds, trying to identify anything that could be a problem. I measure the voltage between the ground that enters the relay board and the ground lead for the crank sensor. There was a difference of 85 to 89 mV with quite a bit of movement This seems pretty high to me so I rewired everything that grounds back at the relay board except crank position sensor ( mat, clt, O2, tps) and grounded them to the same place the relay board is grounded to. Remeasure the voltage difference at 12.3 mV pretty steady. Tried again with all the sensors grounded at the relay board except the O2 sensor and got 13 mV pretty steady.

I have only had 1 sync loss in about 15 minutes run time with the test wiring I did. This is with 3 injectors per cycle and simultaneous. That is a significant improvement over what it has been. I'm going to make the wiring safe and do some test driving to see if this actually eliminates the problem.

So, I guess the o2 sensor (Innovate LC1) was too much current to be grounded through the relay board and MS3. Or maybe it just produces sufficient noise in the ground circuit to be a problem. In hind sight it seems like a stupid plan to ground the o2 sensor through the relay board since the heated sensor is going to have some current draw. It should be grounded back to the main ground.

Or maybe I haven't really fixed it at all. Still need to do some more testing to be sure.

[jsmcortina]

Be sure to follow this:
http://www.msextra.com/doc/general/grounding.html

James

[Kyle]

I've looked at that diagram at least a dozen times and this is the first time I've noticed that the o2 sensor has two grounds. Guess I'm not very observant. I only have three wires coming out of my o2 sensor control box so I just wired them up to the MS. Should not have done that.

Drove the jeep around the neighborhood for about 20 minutes and didn't notice a sync loss while driving. The log shows that I had four during the trip, so I'll try to mess with it a little more. I'm pretty content with it for now.

Thanks for the help
Kyle