Unusual Timing Fluctuations Viewed with Timing Light

[portmod7]

I have an MS-II v3.0, 2.920 code. It is on a Gen-I small block Chevrolet 5.7L. I have also done the MSnS-E Modification. The distributor is a stock GM unit with the small cap, external coil, and 8-pin module.

When looking at my timing with a timing light, it jumps about 8* as I slowly bring the engine speed up with the throttle. At idle it seems fairly steady, but at around 1500 RPM the timing seems to jump around intermittently. Sometimes when it jumps, it's consistent enough that I can determine it's about 8*. At first I thought it could be my timing light, but I noticed that the engine definitely runs differently when I see the timing jump, I am sure the timing is in fact changing. My advance table is very simple and the timing jumps do not correlate with the table or with what Tuner Studio indicates that the ECU is calling for.

When I have 5V connected to the B-pin on my module to allow MS to control the timing, the timing jumps 8* retarded. When I have the B-pin disconnected, the timing jumps 8* advanced.

I have been having this problem for a while and have not been able to figure it out. I've tried changing the ignition settings with no change. Today I got frustrated enough to go buy another distributor and coil, they did not change anything that I was seeing. Would it be reasonable to assume that the issue is somewhere in my ECU? The ECU seems to be able to control timing like it should, but intermittently I am having timing fluctuation problems.

Also, not sure if it's related or not, but once in a while, probably every 20th time I drive the vehicle, I think the ECU loses the tach signal for an instant. It's usually when the engine is cold and seems to have something to do with a specific engine speed around 3k RPM. I used to have this problem much more regularly, that's why I did the MSnS-E modification. When it happens, the engine completely dies for an instant and the vehicle falls on it's face. It almost seems like TPS position has something to do with it too, if I were to hold the throttle constant at that point, the engine will run up to that RPM, die, run up to that RPM again, die, and so on. But if I lift off the throttle for a second and then accelerate up to the RPM where i was having the problem, it usually goes away completely. Its very rare when it happens, I've usually forgotten about it completely and then it will spring up out of nowhere one day.

Any help/ideas would be greatly appreciated!

[kjones6039]

I think a log would be helpful also.

Ken

[portmod7]

Here are some datalogs from over a month ago. I had originally posted them in the thread about my throttle position problem, but nobody had anything to say about them. I think my advance table is different than the tune I posted above, so I have attached the tune that was in place when I got these datalogs. You can actually see the RPM fluctuate with no fluctuation in throttle position, that's when the timing jumps around.

[portmod7]

So I've been thinking, and reading, trying to figure out just how I am going to figure this out.... The coil is what's providing the spark, so what's causing the magnetic field in the coil to collapse at times other than when it should be...

One thing that came to mind is that the only thing plugged into my coil besides the distributor is my tach, so I disconnected it today to see if it changed anything, no luck.

Now I am thinking maybe I could isolate the ECU from the ignition system by disconnecting the 4-wire plug at the distributor and putting a jumper wire in place between the R-pin on the module and the corresponding plug connection so that the ECU still gets a tach reference signal, but can't possibly affect the timing. Is there any reason I shouldn't try that?

If that doesn't show any difference, I was thinking I would make a new 2-wire harness to go between the C and + pins on my module and the + and - terminals on the coil and route them completely away from where the current wiring is routed. After all, the current in those two wires is ultimately what determines when the coil generates it`s spark, right? I was thinking that maybe they get some unwanted voltage from somewhere the way they are routed now.

If that doesn't work, the only other thing I can think of is temporarily running 12v to the coil directly off the positive side of the battery to make sure I am not fighting a bad connection in my power source.

If none of that shows any results, maybe it's time to give it up and go back to the dark ages of carburetors and mechanical/vacuum advance distributors?

[plonker]

A shot in the dark..I have an MS extra that did that...retarded before advancing off tickover....reversed the VR sensor connections and all was good???

John

[portmod7]

Thank you for the idea John, I will throw that on my list of things to try.

[portmod7]

Well, I tried everything I mentioned above, no change at all. I've messed with distributor gear endplay, changed the distributor, changed the coil, removed the connection between the ECU and the distributor except for the tach signal, ran new wires from the + and C pins on the module to the coil, ran a dedicated power to the coil direct from the battery, and changed the location of my timing light clamp on the spark plug wire...

The only thing I can conceive that it can be now is some kind of ground issue or something feeding an unwanted voltage to my module through the distributor body. Is that at all reasonable? I thought maybe it could be something mechanical like a timing chain, but because sending 5V to the B-pin seems to alter the issue, it seems like something mechanical is out?

Maybe I could ground my voltmeter to the negative post of my battery and check the distributor body for any voltage? I am terrible with electricity, not really sure how else to go about diagnosing it? Maybe fuel injection wasn't the best project for me.

[Matt Cramer]

Besides a reverse wired VR sensor, what about timing chain stretch?

[portmod7]

This motor is a practically brand new chevy crate motor with less than 20k on it easy. I suppose anything is possible though. Know of an easy way to check without pulling the timing cover? I figure I could move the crank back and forth by hand and see how much crank movement it takes to move the distributor? Although I know there will be some slop in there, it might be hard to tell if what I am seeing is ok or not?

I forgot to mention above, at one point I mistakenly had the + and C leads from my module to my coil swapped during my experiments and the motor wouldn't even fire. Is that what you mean, or actually swapping the wires from the VR to the module?

Thank you for putting some thought to this and offering suggestions. I don't have a clue how to proceed, any ideas at all are helpful.

[plonker]

Yes,actually reversing/swapping the wires from the VR to the module...

John

[portmod7]

I will try that John, thank you!

[Matt Cramer]

If you are running a modification originally developed from the Extra code, some variants of those modifications won't work with B&G 2.920 firmware. Can you post the full details of how the board has been modified?

[portmod7]

I swapped the VR sensor wires tonight, I'm not really sure what to make of the results.... The timing retards smoothly as RPM increases as if the computer is controlling it, but it retards whether or not I have the 5V connected to the B-pin. Before, the timing would advance when the computer would call for it like you would expect, and wouldn't change at all with the B-pin disconnected except for the problems this thread is about.

There is still some timing fluctuation at ~1500+ RPM, but it's very bouncy, it doesn't stay at the altered timing long enough to even tell if it's advanced or retarded, and it doesn't seem to be a consistent amount judging by the long and short marks on my balancer. So even though the timing is still bouncing around, it seems to be a little different with the VR wires switched.

As far as modifications go.... Yikes.... I'm not even real sure what you would call a "modification", and I put the board together several years ago so it's not real fresh in my memory..... BUT, the only thing I can think of that is not something that was right out of the general V3.0 Board Assembly Megamanual directions is the MSnS-E modification. I can't even remember what resistor I replaced R12 with... I think it was a 1K, but I can't be sure.

Also, I only changed to 2.920 code recently to see if that would fix the problem I am having. I was running 2.905 before that with the same issues.

Thanks again guys. Maybe something from the results of switching the VR wires will tell you something....

[portmod7]

Tinkering with this thing again today.... Fixated on the VR sensor, I decided I would twist and shield the wires to see if it changed anything... I twisted them, wrapped them in tin foil, then wrapped that with a lead that I could ground, and put heat shrink around the whole thing. Grounded it with one of the bolts that attaches the module under the cap. No idea if that was a dumb idea or not, but it didn't change anything.

I still had the VR wires swapped when I started today, and it definitely causes the timing to retard as RPM increases. After reading information on the Megamanual website on how to check if you have the polarity right, I tried lining up the VR teeth with the crank sitting at 15* and then fired it up and checked the timing, it was WAY off, showing like 45 or 50 degrees advanced. So, I am pretty sure they were hooked up correctly to begin with. The timing retard must be caused by the flatter zero crossing when the polarity is reversed? The signal isn't clean enough to produce constant timing?

Swapped the VR wires back and it went back to the same behavior that caused me to start this thread.

So, I ran my coil wire away from everything else, no change. Ran my spark plug wires away from everything, no change.

So, I pulled the distributor cap to try and check the timing chain slop. I can move the crank 5 to 6 degrees before I see movement at the distributor. Also, when I move the crank, it moves 1 to 2 degrees before it feels like I start moving the rest of the engine internals, almost feels like backlash in a rear end gear. I pulled a valve cover to watch the rockers, found one in mid-stroke, and it starts to move at either side of the 1.5* of "crank slop" I described.

Also, if I hold the rotor of the distributor to keep the distributor drive slop out of the equation, it starts to move at either side of the 1.5* of "crank slop".

So... Looks like I have 1-2 degrees of timing chain slop, plus another 4 to 5 degrees of distributor drive slop. Is that abnormal, normal, etc? Am I even making sense? Ha ha....


Thanks guys!

[Matt Cramer]

That's a LOT of slop. Time for a crank trigger?

[portmod7]

I had just about decided that it was ok. All my Google results seemed to indicate that what I was seeing was pretty normal.

Not real sure what to do next. Kinda at my witts end with this thing. Tempted to get an aftermarket distributor if i could ever decide on a brand, but everything I read says the GM ones, like I already have, are the best thing going.

Are there any good ways to go with a crank trigger that won't cost an arm and a leg? Not real excited about re-wiring either.

[kjones6039]

Are there any good ways to go with a crank trigger that won't cost an arm and a leg? Not real excited about re-wiring either.

I have 2 sbc's using 8 1/4 inch, 36-1 crank wheels from DIYAutoTune. I placed mine between the lower pulley and harmonic damper. Take a look at them over at DIYAutoTune. They offer a number of other crank wheel solutions also.

IMHO, you will never regret switching to the crank wheel!

Ken

[portmod7]

After I posted yesterday, I got to researching, and the Ford EDIS method seems to be a pretty sweet way to go, and affordable...

Now I wish I would have opened my eyes and realized it before I went down the road I am on, lol. Just seems like a better way to go all around, makes me wonder why I thought I wanted a distributor in the first place. And if I was going to spend $200-$300 on a new distributor anyway, I think I would much rather put that money toward going to crank trigger.

Is the Ford EDIS what you all would recommend? Is that the setup you run Ken? Is there a better way to go?

Edit: Ken, I see in your signature that you are running a differnt system. Can I ask what led you to that decision?

[kjones6039]

Edit: Ken, I see in your signature that you are running a differnt system. Can I ask what led you to that decision?

So, you caught that, huh?

When I began this project, I actually started with EDIS. The only reason I dumped EDIS was because I ran into issues with rpm at speeds > 6000 rpm. Since I am a weekend racer, that wasn't working for me!

I also wanted to try out the logic level, LS type coils. I am very happy with them, although they are certainly a bit more pricey (by quite a bit) than an EDIS system.

All-in-all, EDIS should meet your needs and your budget.

As a matter of fact, I just happen to have an EDIS8 module, a pair of coils, a brand new set of plug wire connectors and a 36-1 wheel (sbc ready), laying around my shop somewhere, that I would part with very reasonably. Send me a pm if you decide to go that way and are interested in the parts I have.

Ken

[portmod7]

I stumbled on that little tidbit about the RPM earlier, somebody somewhere on the interwebs said the EDIS module will only work below 6250? Like you said, I don't need to go that high, so it would probably work fine for me. I will PM you.


Thanks!