Randy,

You are also correct. I can somehow remember that trivia but not the names of those I met yesterday.

Did you see Jack Benny’s 100th birthday on NBC? On LIVE TV, he said the four stages of age are forgetting people’s names, forgetting their faces, forgetting to zip up, and forgetting to zip down. I’m about stage three and hope to make the century mark with my mind as sharp as his.

Best,
Grady

Grady,
Benny's 100th birthday would have been in Feb 1994, is that what you are referring to? He died in 1974 AFAIK.
I must be in stage 5 - because I am so confused by your post...;-)

The only one still lucid at that age was George Burns, still in stage 2.66 at 100, IMHO.

Here is an update for folks with tach problems ... and otherwise fully functioning CDI unit, i.e., engine starts and runs fine, but tach is dead ... tach works with other CDI unit or on the bench with ramp or function generator input signal:

If no TD signal is detected with AC Voltmeter, oscilloscope, CMOS logic probe, or DROID [a form of logic probe/monitor for EFI system troubleshooting] ... with TD lead still connected to powered tach [and speed relay disconnected for good measure] ...

If the TD lead was connected to a powered-up tach at the time, and the Voltmeter was set to a 10 - 20 Volt AC range, then, yes, it does sound like D8 is blown. Luckily, that is an easy diode to find, a 1N4148! Unfortunately, 15 wires have to be unsoldered to get the board out for a proper repair. A quick fix would be to cut the leads of D8 close to the glass, and solder a replacement to the remainder of the leads, observing polarity.

Just for clarification ... in the following pic, the thick band end of diode D8 goes to the right, the circled hole indicating the Cathode end. And for those wanting the color code ... it is Yellow, Brown, Yellow, Gray ... starting at the thick band end ... 1N4148.

http://forums.pelicanparts.com/uploa...1141408623.jpg

comments on component replacement.
 

Again I am grateful to Early_S_Man for providing the schematics of the 6-pin CDI.

I have a few comments to provide on the recommendations attached to the part list given below the schematics :

It is written about ZD1 : "1N3003B 82 V, 10 W, 5% is OEM but not recommended ! A 45-50 V 1N2994, -95, or -96 replacement is suggested for better protection of T1..."

I disagree. This would have a direct consequence on the output voltage of the converter that would be reduced accordingly. Furthermore the power consumption of the unit would also increase and ZD1 would probably be overloaded.

I would like also to draw attention on another factor to be taken into account when repairing earlier versions of the CDI using ZX82 for ZD1.
The correct equivalent for ZX82 is not 1N3003B but 1N3003BR (cathode-to-case instead of anode-to-case), otherwise the mounting of the zener diode would be different.
I suppose (but I do not know) that the mounting have been changed for implementing 1N3003B as OEM.

I hope this might be of some help.

I am troubleshooting/repairing a 6-pin box, and as of now I have replaced the SCR, Z1, and T1. The box is now making the characteristic whine (it wasn't before I started), but still not producing any spark. The components on the board all appear good (no visible issues) and I have tested most of the componets. I'm left with replacing C8, and I want to make sure that I was looking at a proper replacement, since I've only found radial, drop-style caps at the 1.5uF, 600V.

Any insight?

Thanks,

Pat K
87 930

About the capacitor a polyester 1.5uF 600 V is OK, but should preferably be of cylindric shape with axial leads and should be firmly attached in the box.
But the replacement of C8 should be considered only if it is either open or short-circuited. Did you check this ? Did you check the voltage between D7/C8 and ground ? (about 460 V DC).
The operation of the Schmidt trigger (T2 and T3) should be checked, but this is impossible without an oscilloscope.

I'll check that voltage tonight. If T2 and/or T3 are suspect, it would be fairly easy and cheap (~$1 each) to replace. I will order some along with the cap, if needed.

Thanks!

Pat K
87 930

No voltage at C8 or D7 on this box. I have another Bosch box that has ~400v at C8/D7.

Check for high AC Voltage at 'E' terminal of Tr winding 2 and anode of D6, then check D6 and D7.

E terminal has ~11 V AC.... same as anode of D6.

My supposed good box has ~180V AC at these points.

Thanks!

Pat K
87 930

Dear all
I have a lot of noise on the Schmitt Trigger on point Base T2. I assume it causes misfire. Can that be?

How do I evaluate The correct value for R5. selected resistor? The owner before placed 50k resistor at R5/R6

I am using a 2 V peak to peak sinus signal a a trigger. I s that appropiate.
If I increase that value to 8V pp the box works better due to a better signal to noise ratio.

How do I reduce the noise on the trigger ?

Regards Bob

Triggering signal
 

This is caused by the high voltage converter generating a "noise" at about 3 kHz and this is normal. This is normally without consequence if the slope of the input signal if not too slow.

The R5 value is not to be modified. The R6 value is determined at factory to set the triggering voltage at the zero crossing of the negative slope of the input signal (a bipolar pulse)

This is not appropriate as the slope of such a sine wave is not fast enough not to be disturbed by the 3 kHz spurious signal.
See the attached oscilloscope picture of a simulated appropriate signal showing the triggering signal at the gate of the SCR at the correct time.

http://forums.pelicanparts.com/uploa...1196719836.jpg

A. "This is normally without consequence if the slope of the input signal if not too slow."

B. "The R5 value is not to be modified. The R6 value is determined at factory to set the triggering voltage at the zero crossing of the negative slope of the input signal (a bipolar pulse)"

C. "I am using a 2 V peak to peak sinus signal a a trigger. I s that appropiate.
If I increase that value to 8V pp the box works better due to a better signal to noise ratio.
This is not appropriate as the slope of such a sine wave is not fast enough not to be disturbed by the 3 kHz spurious signal."

What is all this nonsequitur stuff?????????

A. The purpose of using the Schmitt trigger is to make the input NOT
sensitive to the slope of the input waveform.

B. The triggering level is NOT set for zero crossing or it WOULD BE sensitive to
noise. The level is set to about 100mv.

C. A sine wave is MORE than appropriate as a trigger source for testing which
ALLOWS the tester to determine the input sensitivity.

Anymore guessing????????

Sorry. This is -your- "nonsequitur stuff". You are mixing original post with my answers.

I agree, but this is not the point. The point is the instability in the triggering time if some noise added when the slope is too slow

When the pick-up coil is connected there is no more noise by fixing the triggering level at 0 V rather than at 0.1 V

For this simple test I would agree, but the instability in the triggering time caused by the 3kHz noise on a simple sine wave signal is very disturbing for the observation of subsequent signals with an oscilloscope.

Anymore non appropriate criticisms????????

"When the pick-up coil is connected there is no more noise by fixing the triggering level at 0 V rather than at 0.1 V"

Not correct. Remember the pick-up coil is an inductive device (that's affected by dv/dt - noise),
and it's in a noisy environment (the HV distributor) and the green wire from the distributor
is fairly long, so having a trigger point of 0 volts versus 100mv makes the input MORE suspectible
to noise which does exist. That's why Bosch set the input to a value NOT equal to 0 volts.
Use a scope and measure it.

"For this simple test I would agree, but the instability in the triggering time caused by the 3kHz noise on a simple sine wave signal is very disturbing for the observation of subsequent signals with an oscilloscope."

So, then get a better signal generator AND use a better hookup method which will
allow signals as low as 100HZ for triggering.

I continue to disagree about the noise, but I have the feeling that we are not talking about the same thing. I am talking about the voltage in the input signal at which the triggering of the SCR occurs.
May be you are talking about the static voltage measured at the coil input (pin 7, when the engine is not running).
Could you provide us with a copy of the corresponding page of the possible Bosch service manual mentioning these 100 mV ?

"For this simple test I would agree, but the instability in the triggering time caused by the 3kHz noise on a simple sine wave signal is very disturbing for the observation of subsequent signals with an oscilloscope."

This is not a question of signal generator quality or "hookup method".
Sorry, but I prefer my method, more comfortable. Did not you like the simulated coil signal of my oscilloscope picture ?

Loren is talking about the input signal to pin 7 of the CDI box. I ran a quick SPICE simulation and it confirms his statements: The Schmitt-Trigger does not seem to care about dV/dt. In the simulation even a low-frequency sine wave works perfectly all right. The SCR gets triggered slightly after the zero crossing. This puts the level to about -200mV. Changing R6 shifts that point a little bit but not to zero. And it is amazing how accurate the simulation shows waveforms at the gate of the SCR that match real scope traces.

When I use my little square wave generator (555) I can trigger the 6-pin units down to 60Hz with no issue. 2V p-p symmetric around 0V. The 3-4kHz DC/DC converter noise results in some interesting harmonics. Tiny changes in supply voltage change the time when the DC/DC kicks in after the SCR turns off. This causes a slight instability of some scope traces in continuous mode. When I tweak the supply voltage there are audible changes in the spark noise for lack of better description. However, even a simple digital storage scope can capture waveforms with no problem. And with smart trigger options on better scopes things get even easier since you can discriminate the trigger conditions by pulse width. I posted some data from my 4-channel Tek TDS2014 on another thread.

As a hint: The frequency of the DC/DC converter is a function of load and supply voltage and the load is not constant between sparks. You can use the SCR gate signal as trigger source even when looking at the DC/DC converter.

Ingo

Thanks for your interesting comments. You wrote :
In fact some misunderstanding occurred in my previous discussion with Loren. A Schmitt-trigger does not care about dV/dt. I have no doubt about that ! No doubt either about the use of a simple sine wave signal. I was just trying to explain the time instability in the firing of the SCR observed by HKZ Bob in relation with variation in the input sinusoidal signal amplitude.
Do we have to understand that R6 is set (at manufacturing) to have a static voltage of -100 mV at pin 7 ? (with a coil connected).

jmchrist, I don't see how anything connected to pin 7 under static conditions (coil, resistor) could produce negative bias of 100mV. The circuit does not provide any negative voltage relative to ground.

What I am saying is that the trigger happens at a negative voltage. See the SPICE output examples:

http://forums.pelicanparts.com/uploa...1196897821.jpg

http://forums.pelicanparts.com/uploa...1196897832.jpg

The value of R6 is between 10k (brown) and 51k (yellow) with 22k (green) as standard value. You can see how the trigger pulse gets delayed as R6 gets smaller.

Ingo