Electronic Ignition Waveform -- abnormal?

[kenz240z]

Al, I'm kinda in the same spot as a BSEE. Started with a local telco after college doing translations so I got rusty on the EE stuff. I've been a RF engineer for 12 years now. I tend to work in the frequency domain these days...

Sarah, have you tried using a timing light as Al mentioned? As a test, you might connect the timing light to the coil wire to check for irregularities. With respect to the injectors, you might consider getting a noid light to verify they are firing properly.

[FastWoman]

Hi Al and Kenny,

Interesting about the cross-firing. I'd never heard of that. Did you see sparking between the wires? I can't imagine there would be a strong enough inductive effect, even though the wires are ferrite core and spiral wound.

I had a set of translucent NGKs on my old '75, and they were the best wires I ever had on any engine. I think the wires are all opaque blue now (?), and I think the ferrite core is a relatively new thing. Anyway, I have no opinions yet on the newer wires. Everyone (else) seems to like them, though. Besides that, if I go with the MSD ignition module eventually, MSD says their equipment is incompatible with carbon core wires and requires the spiral wound type.

In answer to your question, Kenny, I did test the individual wires with the timing light the same way Al did. I did get triggering irregularities on all wires that looked somewhat like bursts of "fluttering," so that makes a second device (besides my tach) that doesn't like what's going on with the ignition.

Al, I sure do wish I still had one of my old Tektronix scopes. Those things were bulletproof. Unfortunately the ones I had were so old that the caps were all dried up and unreliable. I'm in awe of the capabilities of this cheapo Chinese oscilloscope I bought off of Ebay, which includes including some rather powerful real-time signal analysis tools that would give our old $100,000+ equipment bench (shared between labs) some pretty stiff competition. What the consumer oscilloscope lacks in volt scale accuracy, it more than makes up for in time scale accuracy, convenience, integration, versatility, and computation speed. Honestly, if I had the availability of this device back in the late 80's, I'd have finished my dissertation at least a year earlier. I just need to devise a better probe for using it on automotive applications. I really have to be careful about overvoltages, particularly when the voltages in question are largely unknown, and the consequences of errors extend beyond the price of the oscilloscope to the integrity of our laptop computer!

I frankly find all this high voltage stuff daunting. My domain was uA, uV, and tens of Mohms -- electrical currents through glass electrodes of maybe 10 microns in dia, recorded from neurons a quarter that diameter. I rarely dealt with voltages higher than 15VDC.

[Maples71240Z]

I have been following this thread with great interest. I have a Tenma lab scope that I have used for other automotive apps but not secondary ignition. Is there a reason that a inductive pickup from a timing light will not work?

Edited October 30, 2009 by Maples71240Z

[240Z-Fan]

.. Did you see sparking between the wires? I can't imagine there would be a strong enough inductive effect, even though the wires are ferrite core and spiral wound.

I didn't see any sparking between wires, but I didn't think to check it at night either. The issue was 100% repeatable. The idle would smooth out and the timing light would return to a nice rhythmic pattern. I didn't have a jumpy tack but this was also in a 73z with points.

I replaced the wires with a set of Autolites and the problem went away.

Come to think of it, I had problems with two sets of wires on Honda's. My Honda would barely run in the rain; I didn't suspect the wires because they were new. So I replaced everything else with no success. Replacing the wires on a whim solved that problem.

The second Honda, my sisters, burned a valve for no apparent reason. (I had blamed her for not running high enough octane fuel because it was always pinging). When I rebuilt the engine, I couldn't get it to stop pinging no matter how carefully I set the ignition and cam timing (with 89 octane fuel). I finally borrowed the new wires from my Honda and put it on hers and the pinging stopped. Putting her wires back on brought the pinging back. A new set of Bosch wires got rid of her pinging permanently.

Regarding the cross firing: I had always learned to have wires cross at right angles to avoid cross firing, so I was shocked to see a girlfriends Scirocco had the wires running together in a channel. Since them I have seen many other cars with this configuration. I suppose plug wires are much better than they used to be.

Al

Edited October 30, 2009 by 240Z-Fan

[FastWoman]

Hi Al,

Interesting about the crossfiring. I googled it and found a small bit of info. Apparently it *is* an inductive phenomenon. The wire clips we have on our Zs are similar to other wire clips I've seen that are supposedly designed to space wires far enough to prevent crossfiring. There's a floating triple clip that I think is used for wires 4, 5, and 6 as they loop over the righthand side of the engine. Maybe that would have prevented your 5/6 crossfire problem? Supposedly you can see a corona in the dark around wires that might be crossfiring; however, that wouldn't seem consistent with an inductive mechanism. I'll have to give that a look nevertheless. I wonder if there's a capacitive component to crossfiring, in addition to the inductive. That would seem the more likely cause to me, as current flow is not actually that great.

[edited out "bright idea" that didn't appear all that bright on reconsideration ]

Edited October 30, 2009 by FastWoman

[FastWoman]

Update:

I got the MSD Blaster 2 coil and installed it. The engine started normally, but really missed badly when revving the engine until it warmed up for a couple of minutes (might just be coincidence, as it can be moody the first couple of minutes anyway). However, once I got past that point, the engine revved normally and ran smoothly. It continued running smoothly until the engine reached normal operating temperature, and then it started idling a bit more roughly, as with the old coil.

Here's the interesting thing, though: The tachometer no longer has random spasms at idle and crazy swings on acceleration. Thus, I seem to have eliminated the high voltage ringing on the back end of the burn time. I don't know anything about the current waveform, as I haven't scoped it yet. I'll leave that for tomorrow afternoon, after I have the car fitted with new tires.

I'm guessing I didn't actually solve anything, but simply tweaked enough of the operating parameters that I've diminished the ringing enough not to double-trigger the tach. I'm still betting my control module has gone funky. I'll know more tomorrow.

[FastWoman]

OK, here we go...

I'm a whole lot closer with the new coil:

Peak voltage=500V

Burn time=1.6 ms

I did multiple captures, and this is about as normal as the waveform gets. Some waveforms took negative swings of -200 to -300 V. Some have a much shorter burn time. There's all variety of randomness from pulse to pulse, including large variations in burn time and waveform. I spoke too soon about the false triggering of the tach disappearing. It still false trigger on hard acceleration, albeit not nearly so much.

I also scoped the +12V rail to see if I have some problem with the alternator. Although it doesn't look like it in the above trace, the voltage is mostly quiet. I get the same low level quavering, but I see it at all scales in about the same magnitude on the screen (i.e. not of the same absolute voltage). It seems to be internal to the oscilloscope, relating mostly to the fact I've not been thorough in grounding the oscilloscope to the computer.

What I did see was approx. 16V spikes over baseline (i.e. 30V peak, with respect to ground) from the ignition spikes, measured at my MaxiFuse block. I'll be replacing the condensor off the + post of the ignition coil. I checked at a local auto parts store, and they don't have the Nissan-style condensor (with two wires with bullet connectors). I'm guessing these noise suppression condensors are all approx the same, since values aren't specified? I have a new condensor off of my last alternator, which I can tip with a bullet connector and bolt to ground. I'll try that.

Anyway, although the waveform looks much better, it's still very irratic. My next step, I think, will be to replace the ignition module. Then everything will be fresh and internally compatible:

new/rebuilt OEM distributor and reluctor

rotor and cap apparently very recent

new NGK wires and plugs

MSD Blaster 2 coil

and finally...

MSD 6A ignition module

[alhbln]

(Reviving an old thread)

The original scope trace does indeed look like there is a second trigger, ringing would look different. What you see on the primary side of the coil is the inductive kickback of the magnetic field breaking down, in a healthy coil it should look quite identical to the secondary curve. I would suspect one of the primary transistor drivers or a charge timing capacitor being the issue, you might want to compare the input signal to the ignition and the trigger signal to the base of the power transistor using two channels of your scope.

The scope picture you have posted last is missing the burn phase, also the initial HV peak does look a bit strange having a voltage breakdown right in the middle. Looks like you might have measured the ignition without a plug and engine compression.

You can measure the secondary impulse by creating a simple inductive clamp using a round paper clamp (4 cm width is a good size) which is wired to the shielded wire in an RG-58 coaxial cable, the shield of the coaxial cable then goes to ground via a spring clip. Crimp a BNC barrel connector to the other side for connection to your scope. The coaxial cable should not be longer than 2 meters. 1 Volt is roughly around 10KV, depending on the paper clamps size

(If you need to measure the secondary voltage and signal more precisely, a Tektronix P6015/A HV probe does the trick, if you can borrow one).

If you are in a hurry, just wire some lead solder 5 times around the spark wire (with a gap of 2-3 mm between windings) some inches before the plug and connect the scopes probe clamp to one side of your homemade inductive lead solder pickup and the probes ground clip to the cars ground.

[alhbln]

P.s. carbon spark cables can break and then do funny things to the ignition impulse as the gap acts as a capacitor. Would recommend going for NGK resistor spark wires as example, which use an insulated core with a thin wire wrapped around it. MSD and other brands produce the same type of spark wire using different trade names.

http://www.ngksparkplugs.com/images/prod_images/display/wireset_illustration.gif

[Curtis Snow]

My name is Curtis Snow,

I have a 1978 280Z and I just replaced my alternator with a remanufactured 60amp one. I had Autozone test the old one and the alternator failed their test stand. I cannot get the new tested alternator to show more than 11.5 volts when the engine is running. The battery shows 12.65 volts. I know the voltage regulator is located inside the alternator. Question: What is the T (duel wire) plug used for? The plastic T connector that plugs into the back of the alternator. Where does the wires come from before the plug? I made sure all the connections are clean, including the fusible links and all the wires show continuity. The wires show battery voltage all the way to the positive lead on the alternator. I checked the ground cables as well. The battery is a year old and has been maintained over the winter. I tested the battery at autozone and it tested good. I went back to get a second alternator, even though the first one tested good at Autozone. I now am the proud owner of two good remanufactured alternators and both do not show a voltage from the positive lead on the alternator when installed. Can anyone help?

Thanks

[sblake01]

Check your fusible links. The front one closest to the box. Should be a black 80 amp link. If it's bad, you won't get a charge signal.

[FastWoman]

Hi Adrian,

Thanks! Your post leads me to feel that I ultimately took the right direction: I retrofitted my car with a GM HEI-type ignition module to replace my factory IM. It worked great and took care of much of my misfiring and tach twitch problem. I think the remaining issues probably relate to fuel injection. That's been on my "to-do" list for way too long now. It's hard to get excited about working on my car in this horrible weather! Very soon now.... I hope.... ;-)

Peace,

Sarah