Kawasaki KLR Forum banner

Discussing CDI charateristics

14K views 57 replies 9 participants last post by  Damocles 
#1 · (Edited)
EDIT: I moved some posts from the "2019 What did you do to your KLR today" thread because this topic was basically a high-jack to that thread.



I put an oscilloscope on the pickup coil and the ignition coil, looking at the waveforms. Does that count?

I painted the KLR600 engine covers silver, too.
 
#3 · (Edited)
This trace is from the pick-up coil. The distance from the negative spike to the positive spike represents the length of the reluctor bar. I can't remember how long it is or what the central angle is. I think I have that info over in the Gen 2 >Gen 2 ignition thread.


This trace is from the coil. It is the tach signal.


I had to make a couple of cables that would bring the signal out to where I could latch onto it with the oscilloscope. Today I just played around to see if I could get the signal and what it looked like. What I plan on doing is connecting a two-channel scope up and getting these traces displaying well.

The pick-up coil trace should lead the coil trace by some milliseconds that will represent 10° of initial timing advance. As I increase rpm I should be able to see the ignition timing advance and plot theGen 1 ignition advance curve. I'll do the same with the KLR600, though I will be using a KLR250 CDI unit there.
 
#4 ·
I think I have that info over in the Gen 2 >Gen 2 ignition thread.
I don't know what a reluctor bar is… can you point me to that thread?

…getting these traces displaying well.
It looks badass. Having oscilloscopes in a motorcycle dashboard (not to imply this is what you're doing) would be Ultima Farkle!

The pick-up coil trace should lead the coil trace by some milliseconds that will represent 10° of initial timing advance. As I increase rpm I should be able to see the ignition timing advance and plot theGen 1 ignition advance curve. I'll do the same with the KLR600, though I will be using a KLR250 CDI unit there.
Are you planning on tweaking the ignition, or is this more of an exploratory research thing?
 
#9 ·
The KLR ignition (well, the CDI at least, the Gen 2 igniter is a different animal) has neither mechanical nor vacuum advance. The initial timing is fixed at a nominal 10*BTDC by virtue of the placement of the reluctor bar on the rotor.

The timing advance occurs in the CDI unit through some sort of electrickery which I don't yet understand. It may have something to do with proximity, velocity, triggering speeds, thyristors, resistors, or all of the foregoing. But it is definitely electrickery.
 
#10 ·
Tom,

My guess would be that the chart is meant for checking the performance of the CDI and that the two curves represent the acceptable limits, with the measured performance expected to be somewhere between the two curves. A +/- 2 degree tolerance from nominal seen here for the y-axis is what I am used to from aircraft engines, and some additional variation in the rpm axis would be expected since this CDI is likely an analog circuit using RC networks for timing elements.

Martin
 
#11 ·
@chrider

Martin,

That does seem to be a reasonable explanation. A bit odd in presentation; perhaps the associated word chart didn't make it into the manual.

Now, your comment on RC networks interests me. This CDI system has a conventional pick-up (crank position sensor) coil. The reluctor bar on the rotor is in two pieces. The leading edge of the first piece corresponds to 10*BTDC and the second corresponds to 40*BTDC. Those points correspond to the "Fire" and "Full Advance" markings on the rotor. The duration of each piece is 10* with a 10* gap between them.

Given that, can you describe how an RC network would work to cause the advance? I'm a bit befuddled. I understand RC for charging/discharging, but not using it to create an advance curve.
 
#12 ·
Martin,

One suggestion that was offered (as conjecture by a person whose opinion I trust on such matters) is that the pick-up coil's signal becomes stronger as rpm increases. As that signal increases in strength, it reaches the thyristor threshold earlier thus causing an advance in timing.

That does offer an explanation as to why the Gen 2 reluctor bar is so long so perhaps that is how the TCBI on the Gen 2 works. For the older KLR600 CDI, though, I wonder. The KLR600 rotor is too neatly perfect for it to be a progressing threshold sort of thing. The KLR600 has an initial timing of 10*BTDC and a full advance at 40*BTDC, so a total of 30* of advance. The reluctor bars start at 40*BTDC and have a span of 30*BTDC. It is as if edge detection of the reluctor has to perfect and the advance happens by some other means.

I'm trying to imagine an analog circuit that could, based upon frequency, delay triggering the capacitor by up to 30*. With no advance circuitry, the spark would occur at 40*BTDC. The advance is all-in by 3000rpm, so what must happen is that the circuit retards that spark to ~10*BTDC at idle and the advance curve occurs because the circuit is sensing rpm and reducing the retard until it is gone at 3000rpm.

What could it sense and how could that retard the spark? At 1200 rpm there are 20 revolutions per second, or one revolution every .050 seconds, or 50 ms. To retard the leading edge signal by 30 degrees would be to delay it for about 4 ms. How do you do that?
 
#13 ·
Tom,
I am an electrical engineer, but I've not given CDI circuits any thought before, and while I could come up with some electronic mechanisms, not sure they would be practical. Usually when I start with something I don't know anything about, I look through the internet for ideas. There may be very simple design ideas for this particular problem, but I have no idea. I've spent only a few minutes looking into this, so this might all be FOS (Full of Shit), but here are a few observations:

Glancing at your chart and accounting for the +/- 2 deg angular tolerance, there seems to be about a +/- 10% tolerance in the RPM axis. That would be a reasonable tolerance for an analog circuit that uses a capacitor based circuit to establish some time/rpm reference. I don't think using the crank signal amplitude for that would provide this kind of accuracy over the life of the system, and changes in pickup coil performance would directly affect the spark advance, something I'd be concerned about.

The first few home-brew circuits I found all used PIC microcontrollers to figure out the advance based on RPM. I have not yet figured out how typical analog circuits are done for this application.

I also noticed that the pickup signals have a negative pulse associated with the leading edge of the reluctor, and a positive pulse with the trailing edge, or vice versa. Your oscilloscope shot shows that too. One circuit I looked at used the leading edge negative pulse as input into the (microcontroller based) timing circuit, and the trailing edge positive pulse to trigger a spark at 0deg TDC to facilitate starting. I don't know whether all ignition systems retard for starting, but in Lycoming engines I played with, accidentally firing at 10deg before TDC during cranking typically cracked the starter mount. I think the circuit should fire at TDC or later until RPM information has been established.

I can't invest any time into this right now, but my next step would be to translate the angle vs. rpm curve you got into a time advance vs. rpm curve. That, combined with the location of the reluctor edges available would define what the circuit is supposed to do, and which parts of the curve can be directly derived from the reluctor edges.

Martin
 
#14 ·
Martin,

I cannot say for sure that the CDI is purely analog; I was just looking at a patent that was granted in 1975 to a couple of Japanese fellas that seemed to be mostly digital. The 555 was around then and their design has flip-flops and gates, so why not the Denso hardware 10 years later?

The Kawasaki schematics are no help, they simply have an AMWO* I-O box labeled "Ignition Timing Control Circuit".

What I am going to do is worry less about how it does it and concentrate on what it does. That was what I was originally after; the circuitry is a squirrel that ran by.

It may take me another week or so to get around to figuring out the Gen 1 advance curve. After that, it may be another week before I can get the KLR600 ignition installed.

*A Miracle Will Occur, usually reserved for vaporware or stuff not understood
 
  • Like
Reactions: chrider
#15 ·
I pulled all the CDI posts out of the "What did you do to your KLR today" thread and put them in their own thread; this topic didn't belong there.
 
  • Like
Reactions: chrider
#16 ·
This is really neat, I actually took an old CDI apart once, I think I still have it lying around, although I believe it was a Chinese knockoff. It was mostly digital, the only analog on the board seemed to be for power conditioning and output boosting. It actually had an MCU that I figured probably didnt have write protection on it, and originally I thought about trying to dump the FW in order to create my own somewhat tune-able CDI box. Although I never got that far unfortunately, since another chinese knockoff CDI was only $30 or something and OEM one was like over $130.
 
#17 ·
It's raining cats and dogs. The hills are sliding down the burn areas. News is boring. No car chases in LA today.

What to do, what to do?

I know! Weird Science from the Shop of Horrors!

I'm still looking at data to see if I can develop a curve, but this was too much fun not to share.
 
#19 ·
Still better. The ignition was dumping noise onto the trigger line so I changed the scope trigger to the ignition line so the trace wouldn't jump.
 
  • Like
Reactions: chrider
#21 ·
Sure. I'm using a Pico Technologies two-channel USB scope.

The blue trace, of trapezoidal shape, is the signal from the pick-up coil. That is the coil that is adjacent to the rotor and which reads the reluctor bar. It creates a signal that it sends to the CDI, telling it to dump the energy in the capacitor. That, through the ignition coil, creates the spark at the spark plug.

The red trace is being picked up at the tach signal off of the ignition coil. It represents the ignition spark.

I recorded both of these traces earlier on my single-channel pocket scope, seen a few posts above. I did that to see what they look like and what their relative magnitudes are. The pocket scope is good for quick and dirty stuff as it is so easy to fool with, but there is only one channel. With the rain, I've got a good excuse to waste time and not do productive work like picking weeds and mowing the lawn, so I've gotten around to building some test cables to grab signal with and hooking up the Pico.

What we're seeing is the spark, which is supposed to be at 10*BTDC at idle, advancing to what is supposed to be 40*BTDC as I increase the engine speed up to about 4000rpm. It's hard to figure out what the advance curve is from this video because the time scale is in milliseconds and the number of milliseconds that corresponds to a certain number of degrees of crank rotation changes as the rpm increases.

In the first two videos, I had the scope trigger set on the blue trace, trying to set it just before the peak of the trapezoid. I wanted to measure the spark advance from there. The ignition was putting noise on the blue trace and carrying the trigger point along with the red trace, so the output jumped when the noise on the blue line decayed away enough. I changed the trigger to the ignition, which has a negative spike I could rely on (the blue trace is all positive), so we watch the blue trace retard away from the red trace instead of the red trace advancing away from the blue trace. Mathematically the same.

I have taken some readings at idle, 1000, 1500, 2000, 2500, 3000, 3500, and 4000 rpm to try and figure out what the advance curve is. The scope has rulers that can read the output in degrees, but it's a PITA. I have numbers to crunch.

By the way, @pdwestman, I have tried to see the effect of the Y/R wire, energized and not. The most I could see was that it made the spark signal slightly jittery. I could not see any spark retard. I'll have to take the time scale down into the microseconds ranges to see if there is any retard. I still think it just defeats an anti-kickback function.
 
#22 ·
Variables are my ability to capture traces at the specified RPM and my ability to interpolate the scope timescale. The curve is pretty linear, though, so it seems like I did OK.

 
  • Like
Reactions: Damocles
#25 ·
Photos are allowed after 15 total postings, normally. But some computer wizzes have successfully posted pics at post #1!
It is an anti-spamming measure. Many go to the "TEST POST" thread and burn thru the 15 required postings.

But I cannot even remember where to find the "TEST POST" thread when I'd like to 'link it'!
@Tom Schmitz might help, if we can get his attention?
 
#26 ·
#28 ·
Here are the promise, i weld 8 mm to the magneto tab (pista),or whatever you named, the electrĂłnic lĂ­mit for klr 650 is 30 degress yo protect engine blow so i made It 38 degrees with strobe lamp and runs like hell to 8,000 rpm ,im afraid to spitout the pistĂłn from the exaust
 

Attachments

#29 ·
You can ser the diference in the ignitiĂłn timing at high revs. avance degrees
 

Attachments

#31 ·
And here is the adaptación gen1 ,with ,maneto and eléctrical wire of gen2 ,with Chinese CDI (d.c.) from 62 mm magneto tab( pista) to 30_32depending the degradess Rev you want , i put 32mm large
 

Attachments

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top