Testing ignition components: theoretical question

[jtexas]

This is a purely theoretical question right now - I'd appreciate whatever insight anyone might care to share.

Testing the stator and power packs of a CD ignition requires measuring peak voltage. According to my (very) limited understanding of AC voltage, my DMM probably measures "average" voltage which is by definition 63.7% of peak (unless it computes RMS voltage in which case it will read 70.71% of peak). I gather that these values will vary if the AC output isn't a sine wave (which it probably isn't), but nevertheless can I not approximate peak voltage by dividing my DMM reading by .637 (or .7071 as the case may be)?

Example: DVA output on the orange wires leading from the power pack to the ignition coils should be at least 150V. If I measure an average AC voltage of 96, can I call it good?

Thanks!

 

[Laddies]

Re: Testing ignition components: theoretical question

John, I am not smart enough to answer that question but it sounds right in theory. If you want a answer the best advice I could give you would be to call the service line at Rapair, they are the most knowledgeable electronics people in the country and have always bend over backward helping us with any questions we have had--Bob

 

[eli_lilly]

Re: Testing ignition components: theoretical question


From what I understand, the ignition system pulse is so quick and the period between pulses is so long that a regular DMM won't properly perform measurements on the system. The adapters supposedly elongate the pulses, I suspect it wouldn't be too hard to build one with a little research (perhaps some type of RC circuit).

-E

 

[Chris1956]

Re: Testing ignition components: theoretical question

I had always thought you would need an ocilloscope to see the peaks of the voltage, since a VOM would not respond quickly enough.

 

[jtexas]

Re: Testing ignition components: theoretical question

Theoretically the pulse frequency of a 3-cyl 2-stroke at 1200 rpm would be 60hz...only one way to find out whether I can measure that...

It's not like I'm the first person to ever think of it, yet nobody anywhere that I can find is suggesting it...so no doubt it doesn't work that way...just really curious why, is all.

chris they do make meters that measure peak, also adapters for multimeters, but they probably don't work with the economy models.

 

[Paul Moir]

Re: Testing ignition components: theoretical question

The 63% figure depends on the signal you're measuring being a perfect sinusoid. A 50% duty cycle square wave would be 50% for example. Fortunately the output of a stator is (normally) a regular sine wave.
(EDIT: I see you have that covered)

The frequency is the big trouble. Most multimeters block lower (cranking type) frequencies as part of their DC blocking when you use them in AC mode. Most usually they've got declared accuracy from about 50Hz through 400Hz, but you should consult your manual to be sure. 1200 RPMS would be fine if the outboard was running naturally.

As an example of the sort of errors you can expect to get, have a look about 1/2 way down through this post:
http://www.iboats.com/bbboard/bbBoard.cgi?a=viewthread;fid=17;gtid=91991;gpid=92010#gpid92010

On the other hand, it's very easy to build a peak reader adapter. Just a bridge rectifier and a capacitor - Radio Shack stuff.

 

[jtexas]

Re: Testing ignition components: theoretical question

This is great stuff, Paul, but I'm pretty much a hacker when it comes to electronics...I searched unsuccessfully for a schematic...so now I'm guessing...

I connect probes to the rectifier inputs and the DMM leads to the rectifier outputs?

The capacitor goes across the rectifier outputs?

DMM set to DCV?

But the big mystery (to me): how to determine the required capacity of the capacitor ????

Thanks!

 

[Texasmark]

Re: Testing ignition components: theoretical question

Capacitor energy is 1/2 C(V squared) for energy content measured in "joules" (watt-seconds). Watt-seconds x occurences per second = watts (heat content).
(C is capacitance in microfarads, and V is in volts)

If you are talking about a "snubber cap" across a diode, you are talking about a dinkey dude (.01 uf typically) to snub the leakage spike that gets through the junction capacitance of the rectifier.

Most multimeters measure the voltage that is equivalent to the DC value which is Root-Means-Squared (rms). But that is not "TRUE RMS". True rms measures the real heat content and usually is associated with complex waveforms (like ignition systems) that have peculiar waveshapes.

What else do you want to know?

Mark

 

[jtexas]

Re: Testing ignition components: theoretical question

We're talking about a reservoir capacitor, to smooth the rectifier output. Duty cycle at cranking RPM of 250 for 3-cyl 2-stroke would be what, 1/25 of a second? Not sure...how much current? Output from the stator should be 150V.

And there will be some voltage drop; output from the timer base is only half a volt - would this device measure that?

thanks!

 

[Paul Moir]

Re: Testing ignition components: theoretical question

Ah, a fellow hardware hacker! Just improved my shop with a Tektronix 465 - from a '50s era Heathkit o'scope. I'm all ready for the 1970s now.
The capacitor is for ripple smoothing, just like a large on in a typical transformer power supply. Since the load is very small (10 MOhms for a typical DMM) it doesn't have to be very large. And since we're charging it to a couple hundred volts at times, a small one would be advantageous since it's safter.

Iboats member schematic posted a schematic a while ago in Non-Boating Tech topics:
Click here

You have to add the 1.2v drop but that's not hard. It's certainly a lot simpler than the active (OPAMP based) peak detector I made!

Lots of current is coming from the stator. It should pump up the capacitor quickly.

 

[jtexas]

Re: Testing ignition components: theoretical question

Excellent. I do have some garage sale computers and VCRs in the garage from which I occasionally scavenge components...but it's just a sort of self-taught hobby.

Looks like the negative dc output is shown in red, positive in black? Polarized capacitor? How crucial is that resistor?

And I guess measuring 1/2 volt peak from the timer base is just out of the question?

Thanks!

 

[Paul Moir]

Re: Testing ignition components: theoretical question

Resistor bleeds off the 150v or so on the capacitor. It's not critical, you just watch the voltage drop on your DMM until it's safe if you want.
The way I'm reading his schematic, he's got red as positive. The bars on the diodes are the bars in the -|<- diagram, that is to say they mark the cathodes. But really it's immaterial.

Cap is non-polarized but could be if you had one. I don't see any trouble going to a 1uF capacitor either.

I don't think you'll find those diodes in a VCR, but the high voltage bridge rectifier from the power supply of a modern TV would be just the thing. They're normally about 400-600PIV. The IN4007s are rated at 1000PIV which is really excessive. IN4006 would be just fine.

The 1/2v measurement would be the tricky bit where my active peak detector is actually somewhat worth the trouble. I wonder if you could bias the diodes with a pair of AA cells. Let me get back to you on that.

 

[Texasmark]

Re: Testing ignition components: theoretical question

If you have a capacitor in a battery charging circuit it is for spike suppression (saving your diodes from reverse voltage breakdown), not waveform smoothing like filtering the power supply of a piece of electronic equipment.

The battery is your filter capacitor, for that, if you will, and it doesn't give a hoot about waveshape. All it wants is amps going into it rather than out of it and it will take all shapes since the internal impedance of the battery is milliohms.

(If you can pump 200 starting amps out of a 12v battery and read 11v or so at the terminals, it's internal impedance is milliohms and it will eat whatever amperage you want to feed it.....within reason).

d

Mark