Jumping Tach Signal

[cygnusx1]

Last night I hooked the laptop up to MSII and went for a drive to datalog some stuff.

 

What I noticed is that around 2000 rpms with little to no load, my tach was dropping to 1500 and then back to 2000 intermittently. It happened a few times during the datalog and I was able to see it on the graphs. Nothing else obvious changed other than the RPM jolt. I did feel it in the car like a slight loss of power for a second. The PW stayed dead on and the timing was dead on. My voltage does jump around alot on the datalog. It goes from 13 to 14.5 in spikes all over the place but I can't correlate them to the tach noise. I do have a noise filter on the MSII 12v input.

 

The engine did not change RPMS, just the tach needle and the signal to MSII changed. This tells me it's a signal problem. I use an 83 Turbo Dist. wired to the specs on DIY's site. It is also shielded wire, grounded on the MSII end.

 

What should I check out first? The car runs great otherwise. It only happens at around 2000 rpms with little or no loads. Could it be bad plug wires, plugs or cap/rotor?

[naviathan]

Yeah I would look into the typical tune up items like plugs and wires first. The voltage change could be associated with weak plugs making a harder draw at that point for some reason. That is odd though. Post the datalog?

[z-ya]

Are you talking the tach in Megatune, or your dashboard?

 

Pete

[flames_187]

i noticed this too in my car but i was also running rich (while tuning) and flooded my plugs while tuning the crank...i brought some new plugs and a new fpr. I am going to install my fpr first run for a bit then plugs...I will let you know if any of this helps.

 

P.S. for me it was happening around 1500 and dropping to maybe 1300 will little throttle. i didnt eye it that closely cuz the afr was in the mid 11s

[flames_187]

my problem turns out to be different. I just realized my MAP is reading in the 30s for all rpm...

[cygnusx1]

Are you talking the tach in Megatune, or your dashboard?

 

Pete

 

The stock dash gauge shows a split second dip in the RPM's as does the datalog. It happens VERY infrequently, about once every 30 miles or so. It's as if a few pulses from the dist. are dropping out causing a low RPM reading for a split second. I changed my ignition Predictor Algorithm to "1st derivative" from "1st High RPM/ 2nd Low" and the problem seems to be even less frequent now. I drove it for about 150 miles this weekend and only observed the tach jump once.

 

I am reading about the way the signal is processed and might try playing with Next Pulse Tolerance (%).

 

I don't want to admit but I havent changed the wires, cap plugs yet. They are about 4 years old now but have only about 10K miles on them.

 

Now that I think about it........it cant be the way the software is processing it because the dash gauge is run right off the coil right? Or am I wrong because the coil is fired by MSII.

[Zmanco]

Unless someone rewired it, I'm pretty sure on the 280z the factory tach is triggered off the coil. And I assume you have the coil fired by MS? That would suggest, like you're thinking, that the spark is indeed really dropping out and it's not just a s/w processing issue for the megatune tach gauge.

 

Do you notice any change in the engine when this happens? Or is it too brief to notice? Are you sure it's not a reset?

[cygnusx1]

It's not a reset. I do feel some "deadness" or maybe a change of engine tone during the tach-drop event, but it's always during light throttle, low RPM, so it's not real pronounced. I will post my datalog tonight that shows the events or at least a screen cap of the events. It repeats two or three times in one datalog with an almost identical shape to the RPM curve each time, at the same RPM's.

 

I am going to order some new plugs and wires. I need to custom make them to route them the way I did around the front of the motor. It's a long process. Maybe I'll just pick up a cheap set of standard Z plug wires and try those for testing.

[linzerl]

Have a look at your DeltaT values.. and ensure there are no spikes/drop-outs. DeltaT is the time between ignition triggers.

[cygnusx1]

Thanks, I'll get a look at the Delta T values.

 

I have gotten stuck twice on the road since MS-II. Both times the car just quit with smoke and popping from the tail pipe. When I tried to crank the key for a restart, the tachometer needle would jump, during cranking, from 0-3000-0-4000-5000-6000 RPMS swinging back and forth like an erratic, random, pendulum. I also got smoke out the tailpipe during cranking but no start. I tried cranking with wide open throttle, in flood clear mode and got the same tach behavior but no smoke. Both times, after 25 minutes of sitting, the car fired up and ran perfectly fine.

 

What gives?

Is my VB921 overheating?

Is something else getting hot and modifying the spark timing signal?

 

My coil dwell is usually around 2.6ms under most driving conditions but check out the huge dwell to 7.8ms during cranking, in the begining of the datalog.

 

 

I have attached a Datalog taken a couple of weeks ago that might have caught some tach noise. Experts, take a look and tell me if you see any odd things in the ignition signals.

Ignition Noise-CygnusX1.rar

[Brad-ManQ45]

I wouldn't be surprised if the problem is the optical module in the dizzy - that's where MS gets its' signal from.

 

When these start acting up you get the 25 minute waiting periods to cool down.

 

It is for this reason that I am going MS with hall sensor and magnets - U want NO PART of the stock EFI/Ignition system which has left me stranded numerous times and the dealership richer.

[cygnusx1]

OK now we are getting somewhere! I really know nothing about the ZX Turbo distributor. I am now going to do some research on it and explore the options. Deep in my wallet, I felt the urge to convert to coil-on-plug and dump the dist. I don't yet know anything about, it but you can bet that by the end of next week I will be a pro. There is nothing worse than an intermittent problem that leaves you stranded at it's own will.