[TheMonkey]

i have my new EFI motor on a test stand. i'm trouble shooting why my lambda sensor won't stay on after start up. i found 32 volts of AC voltage off the charge wire from my alternator. ignition system is not from the ECU yet, it is still the MSD system with timing from the distributor.

MSD Pro-Billet Distributor with the VR sensor feeding (green & violet wires) the MSD 6AL box triggering an MSD HVC2 Blaster coil. alternator is newly rebuilt from Napa.

the green & violet wires are not shielded, and next step will be for this motor to trigger from the crank trigger with another VR sensor, but will have a shielded cable.

where the heck is this AC current coming from, and how do i fix it?

thx, Scott.

[george kujanski]

Do you have the alt output connected to the battery system on your test stand or is the out put not connected to anything? You may see unusual things if the alt is not connected to the battery system in the normal fashion.

Are you measuring at the alt output and ground?

George

[TheMonkey]

George-

yes, the alternator is wired up as it will be in the car. the large charge wire and the small red wire (i think the voltage reference wire) both go to a positive lug i put on a main board just behind the motor (think wooden firewall). this positive lug is the main point which feeds power to ignition, gauges, fuel pump, ECU, lambda sensor. the small white wire goes through a small automotive light to the ignition.

i measured from the body of the alternator to the alternator's charge post (post on back where the heavy charge wire connects).

i think i'm headed for a crash course in noise filtering. i've heard about it, but really don't know much about it. i've heard that noise gets cleaned up with capacitors or noise filters.... fuel pump can be source, or bad alternator, or these VR sensors which send an AC signal. as i said, i don't know much about this but would like the 'automotive electrical noise for dummies' lesson. do you have any links where i can sharpen my pencil on this? at least i'm learning my lessons with the motor on the bench instead wiring everything in car already....

i'm going to clean up some of this wiring on the bench. i read through the MSD instructions which suggested i route the power and ground wires directly to the battery, which i violated.

Thanks, Scott.

[BVR421]

Have you checked DC voltage yet?
what prompted you to look at ACv?

[TheMonkey]

Originally Posted by BVR421

Have you checked DC voltage yet?
what prompted you to look at ACv?

DC volts are fine. 12.4ish on the battery, and 14.5ish when running.

i diagnosed that my o2 sensor was not working because of electrical noise. voltage was fine, and it worked fine when motor was not running, but shut down as soon as motor fires up. my neighbor suggested checking for AC current at the alternator. he said when an alt goes bad it can send that off and start wrecking things like batteries and electrical components. i had never heard of that. the support pages at the o2 sensor site suggested the same thing about checking for AC current.

last night i looked at my wiring. i'm a total knucklehead. i did not heed the warnings about wiring from MSD. not only did i not connect the MSD box directly to the battery, i had the VR wires from the distributor wrapped in a couple loops around the ignition system right by some other power wires.

i'll clean this stuff up and see what happens. not sure why MSD doesn't use shielded cables for the VR wires. VR sensors are a capacitive device that sends AC current and voltage gets stronger with faster revs.

i'd sure like to learn about how to find any other electrical noise and how to suppress it. i remember high school in the 80's, a bunch of people drove toyotas from the 70s. all of them used to whine through the radio as the transmission wound up, so you could hear the shift pattern through the speakers.

[Half-Inch Stud]

There is a bad diode on the ALTernator. It's still alternating. Should be Directing.

We all know that, but needing simple type-setting.

[george kujanski]

There is plenty of electrical noise in an automotive system; the MSD box, the ignition system, etc. that could be read by a multimeter. An oscilloscope should be used to determine what the power bus looks like.

Fighting electrical noise requires good grounding; all devices that are grounded in a car should have a good electrical ground back to the battery negative. ALL metal must be electrically connected together and connected back to the battery negative.

Do not connect things like a radio ground and the MSD ground close to each other and then run a ground back to the battery, the radio will undoubtedly see a lot of MSD noise. Run the MSD circuit directly back to the bat negative; you can connect the radio ground to metal locally and then the metal to the bat, for example.

If you look at the wiring on a modern EFI car, the return (ground) connection from the sensors, for example, are run back to the ECM rather than connect to engine metal. That way all the sensors and power are referenced to the ECM local ground rather than grounding the sensors at the engine and the ECM ground at the dash. In a car there may actually be a voltage difference between the engine metal and the ECM ground, and this difference (or noise) will confuse the ECM.

High power audio amplifiers should also have the return (ground) run back to the bat negative rather than local metal.

George

[TheMonkey]

thx guys, i'll report back what i come up with.

in the meantime, i found this fantastic site that is giving me the education i was hoping for. sounds like the guys that fool around with mobile amateur radios are on the front lines of electrical noise. http://www.k0bg.com/noiseid.html

[BVR421]

way long time ago the military has just about every wire in a vehicle covered in grounded metal braid shielding, it looked very "no leak"

[zbuickman]

when you hook up shieled wires also remember to connect the shield drain in one spot only. DO NOT CONNECT("ground") both ends.

Edit: also if you twist pairs of wires at differnt rates(twists per inch) you can run the wires unshielded with nearly no crosstalk(inductance) in a bundle. this only works with retun circiuts though.

[TheMonkey]

i checked the battery voltage on the AC settings with the battery disconnected (engine was running with it connected when it read 32v). disconnected, it read 27v AC in the battery, which we KNOW there is no AC voltage in there.

so, my digital voltmeter is of the cheap variety which does not take a true sampling to read AC voltage, but rather calculates it based on the peaks of the signal and is prone false readings with very little current. or something like that....

anyhow, Stuart has a portable scope if we can hook up, maybe get a true reading and report back.

meanwhile, i think if i clean up my ignition wiring, my lambda sensor should work.

from Wikipedia:

****
True RMS Converter

When measuring the value of an alternating current signal it is often necessary to convert the signal into a direct current signal of equivalent value (known as the RMS, Root Mean square, value). This process can be quite complex (see Root mean square for a detailed mathematical explanation). Most low cost instrumentation and signal converters (for example handheld multimeters of the sort used by maintenance engineers) carry out this conversion by filtering the signal into an average value and applying a correction factor.

The value of the correction factor applied is only correct if the input signal is sinusoidal. The true RMS value is actually proportional to the square-root of the average of the square of the curve, and not to the average of the absolute value of the curve. For any given waveform the ratio of these two averages will be constant and, as most measurements are carried out on what are (nominally) sine waves, the correction factor assumes this waveform; but any distortion or offsets will lead to errors. Although in most cases this produces adequate results, a correct conversion or the measurement of non sine wave values, requires a more complex and costly converter, known as a True RMS converter.
****

[zbuickman]

also if you want to eliminate most of the false readings but dont want a scope invest in a simpson analog 260 series multimeter these work quite well. however for ECM and Elecronic use there Impeadance can be ,well, LOW for cases where this happens a nice DVM that also has freq and duty cycle functions is a bare minimum when working with automotive ECM's My fluke will differentiate between pusled DC and square wave AC sinisoidial AC and sawtooth AC but does not diferentiate between the AC forms but does read them all (including elevated waves it will give +,-,avg and center). and the ultimate is a scopeing meter w/ logging capabilities$$$$$$ and the most acurate is an osciliscope as mentioned before

please excusse my teermanoligy as recall is rather hard tonight for some reason

good luck

[69geeto]

Hard to beat the old Simpson 260 I think you are on the right path though with the scope. That will give you a true reading and weather it is as George mentioned, just noise, or something more destructive. My guess would be your low cost meter and alt. noise.
Also as noted, without having a true RMS voltage, who knows where the conversion might end up.
Let us know what the scope says.

[george kujanski]

The suggestion of a Simpson 260 (or similar analog meter) is a good one. Quite often folks not used to high input impedance meters (even experienced folks) can easily get confused with high, unexpected AC voltage readings due to noise.

Since the 260 has lower input impedance than a digital multimeter, it doesn't respond to such noise.

George

[Stuart]

Scott, I thought of something else you can try to eliminate the alternator as a source of noise - simply disconnect it and run the engine without it, just off the battery. As long as the battery has a good charge it will run for a fair amount of time without the alternator in the circuit.

I suppose you should make sure any of the EFI wiring is kept away from the alternator even when it's disconnected, just in case there's any sort of inductive currents from the rotating assembly.

Also, there was a good article in Car Craft a few years back about Megasquirt EFI - in the article they mentioned they had a problem getting the thing to work right until they found out one of the wires was intermittently shorting out against their equipment.
http://www.carcraft.com/techarticles...ion/index.html

[TheMonkey]

i've pretty much finished the conversion over to the crank trigger, and need to assemble things back together and i can take a look at this noise issue again.

i picked up a cool toy. this PC based scope. www.hobbylab.us it's not super fast, but is perfect for this project and general automotive stuff.

i'm wondering though.... when i get things back together, and i want to take a look for any RFI from the alternator, do i just probe the power output from the alterator against the ground from it's casing? when reading it, what's good, what's bad, and what to do if it's 'bad'?

here's a scope shot from the hobbylab scope. it is from a simulated 36-1 trigger wheel (red trace), and the ECU output to the MSD 6AL box (blue trace - drops to ground to fire). TDC is on the falling edge of the 5th tooth after the gap. in this example, you can see that ignition is getting fired right at 24.1 degrees which foots with what it said it should be doing. anyhow i would like to use this to make sure i don't have RFI that can potentially cause probs. any suggestions?

[Engine-Ear]

does the hobbylab pc scope have actual scope probes???

Adding a low-bandwidth (20 MHz is plenty) analog scope (Tektronix 22xx series) and a Simpson 260 or 270 is money well spent if you are delving into the world of "RFI" and digital circuitry...

as H.I.S. said, I would check for healthy diodes in your alternator, too!!

Good luck with it!

[TheMonkey]

Originally Posted by Engine-Ear

does the hobbylab pc scope have actual scope probes???

...

yes, 2 probes, 2 channels.

that screen shot is actually putting the probes to the tach circuit in the ECU, and the ignition out pin from ECU.

[Engine-Ear]

what do you connect the probes to??

what are the specs on the probes (1X/10X, how my pF load capacitance, operating bandwidth)?

help this gEEk out, will ya?

[george kujanski]

One suggestion I have is to connect approx a 200 ohm,2 Watt resistor from probe tip to probe ground while taking measurements. This will provide a low impedance for radiated noise and if your probe has shielded cable, that should give you a more accurate picture of the AC voltage in your system.

George