Problems getting a spark, CDI fried? Help!

Discussion in 'Electrical' started by Kadupult, Dec 31, 2013.

  1. Kadupult

    Kadupult New Member

    Hello all! I have had some real problems with my electrical components building this 2 stroke, the first stock CDI that came with the kit I received was completely dead on the ohmmeter, wouldn't spark at all. ordered a new one and after replacing the CDI and the magneto, it started for the first time yesterday. But only a mile down the road I stopped getting spark and once again, tested with an ohmmeter, dead CDI:icon_cry:. tested the magneto, its seems fine. What is causing this? Am I frying my CDI? How can I fix this? Thanks!

    EDIT: I would like to also point out that this is a brand new engine, CDI and magneto... So it seems odd to me that just riding the bike less than 3 miles would kill the CDI. I've read on here how unlikely that is, so I feel as though I'm doing something wrong, maybe not grounding correctly?

    Thanks for all the help. I really appreciate all the information on these forums.
    Last edited: Dec 31, 2013

  2. crassius

    crassius Well-Known Member

    would be nice if a CDI tester or a second bike were handy to test it - it may just be a wire that is bad inside its insulation or a connector hanging by one strand or something like that that is a problem with the bike instead of the CDI

    anyone local to you that you can try it with?
  3. Kadupult

    Kadupult New Member

    Sadly no, I'm alone in this one. I don't think the bike did anything to the wires this early on, I've gone less than 5 miles on this bike since this kit has been out of the box. but it's a possibility, I won't rule it out. How could I find that out for sure? As I stated before, the CDI isn't showing up with any reading at all on an ohmmeter, so I'm pretty sure that's the problem. I would also like to know how to avoid this problem in the future in addition to knowing how to fix this. (hopefully I can do SOMETHING besides ordering a third CDI)

    Thanks for the help!
  4. crassius

    crassius Well-Known Member

    unless you get someone from Texas to PM you, you may be stuck

    some weirdness with a bad solder connection in those new wires may be causing it, or I once opened two CDI boxes in a row with bad CDIs in them - no way to test these without building a CDI tester, since many go bad with a fat spark, but wrong timing
  5. Kadupult

    Kadupult New Member

    A fat spark and wrong timing can kill the CDI? interesting, that may have been my problem, I was trying out a high performance three ponged spark plug from BB when this happened. I'll try to strip off some insulation to get a reading today, will post results.
  6. crassius

    crassius Well-Known Member

    no, fat spark doesn't hurt timing, it's just that you need both good spark AND correct timing to run

    I've seen CDIs that had a good spark but wouldn't run the motor - that's why I built my tester with a spark plug so I can see the spark and timing marks so I can see when it fires

    I have also seen a painted motor where the washers under the head bolts were painted giving no ground to the head - spark plug only sparked when you held it on a muffler nut

    if you can hold the plug in your hand and get no 'bite' when you roll the bike and can see a short 'blip' on your meter hooked from blue wire to ground when you roll the bike, then CDI is probably bad
  7. Kadupult

    Kadupult New Member

    Oh I see. well the last thing I heard before the engine died was a sharp snapping sound, but that could have easily been anything, even gravel just hitting the frame. But I convinced myself it was the sound of something breaking.

    I might have to build a tester, I know it was doing alright at first, it was running really well for about 30 minutes and then it suddenly just stopped getting spark. The reason I think it isn't timing, is that I've done spark plug tests and haven't seen a single spark since the moment it died.

    I don't have any paint on the motor, and I know when it was working it would spark just fine against the engine.

    As for stripping the insulation around the wires, did that and not even a flicker on the ohmmeter. It's definitely a dead CDI... but how?!

    I'm going to order another CDI or two in the meantime, I'm starting to really not trust these things... Anyone know of a vendor that sells reliable CDIs? Maybe even a higher quality alternative?

    Any idea of what I could be doing to cause both of my CDIs to die so quickly? Maybe I'm just really unlucky. I would hate to buy a third one, just to have it fry because I've been making a mistake this whole time.
  8. HeadSmess

    HeadSmess Well-Known Member

    ohmmeter? sounds archaic. i hope its actually a multi meter with various functions...such as continuity/diode check!

    you will only get high resistance or open readings across cdi wires, blue and black.

    with diode check, youll have a diode reading one direction, nothing the other way.

    should have a fairly low resistance between the plug lead and the black wire. not silly low, not silly high... a couple of K from memory.

    also check for continuity between plug leads and all bits earthed, such as the head mentioned previously...

    most posters here are agreed that the cdi unit itself rarely dies. the magneto section, all the time.

    snapping sound...are you sure the magnet on the crankshaft is even spinning? someone had a snapped crank here at one point...

    check everything. cant be that hard, theres only two wires!

    dont bother with the multi prong plugs ;) spark still only jumps the one gap. iridium/plat plugs on the other hand...personal choice.
  9. crassius

    crassius Well-Known Member

  10. BigBlue

    BigBlue Active Member

    Here's a tip that I got from Jaguar:

    Disconnect the CDI from the stator coil. Connect an AC voltmeter to the coil wires (blue and black). Push the bike with the spark plug removed and see how much voltage is generated. I have seen as much as 50 volts on mine. Drastically less indicates a bad coil. If the voltage is suficient then all that is left is the CDI/high-voltage-coil, assuming that the spark plug isnt screwed up.

    AKA: BigBlue
  11. Kadupult

    Kadupult New Member

    Thanks for all the help guys.

    Yes, I am using a multi-meter to check the ohms, sorry, that is what I ment.

    The magneto checks out just fine, reading a little high on the multi-meter (like 410-440) but the magnet rotates, I know it isn't on backwards which is something I read happened to someone... it's pretty much a brand new magneto, not even the slightest hint of discoloration.

    Just updating, I'm about to order a new CDI, I have tried every freakin test with this thing and resulted in nothing. I'm hoping it was just my luck and I got 2 bad CDI units, but I might order another magneto just to play it safe in the future... Best vendors for these electrical components anyone?

    Still looking for the root of the problem. My dad just thinks I'm unlucky haha.

    EDIT: Magneto Mystery is a good read, will try some of these tests while I wait for a new CDI.
  12. crassius

    crassius Well-Known Member

    a while ago we got some stators in that were reading around 450ohms

    the really old ones used to read around 250ohms & made bikes a bit hard to start so they started making them around 350ohms (more windings) - when we got the 450s in, we were worried about frying CDI units, so sent them back

    not sure if these extra windings will cook a CDI, but we didn't want to try it

    dig your meter probes in really well to get a clean connection, and if it is still reading high, you might want to swap it for a 350ohm stator
  13. Kadupult

    Kadupult New Member

    Ah! I see, is it just luck of the draw to get a 350ohm stator or 400+?

    Will try getting a cleaner connection in a bit. Thanks.
  14. crassius

    crassius Well-Known Member

    AFAIK, someone got these somewhere near the source of the supply chain thinking they might make a fatter spark - we said 'no thanks' and haven't see one since

    I have no idea what the online sellers have now, but haven't seen one come in for testing or repairs since I built the tester and started putting free CDI & magneto test in my ads..
  15. HeadSmess

    HeadSmess Well-Known Member

    the more winds would make more voltage.

    unfortunately you then have to design the cdi around the new, higher voltage.

    it would run more advanced, simply due to the way the CDI circuit is configured.

    if theres no zener diode or similar way to restrict current through the gate of the SCR... youre gunna fry it, no matter what, and then... the main capacitor needs to be up rated to take the extra voltage. exceed rated voltage, and poof... the magical smoke inside escapes! :)

    then is the higher voltage coil up to the CURRENT demand? a discharged cap is much like a short circuit...current goes through the roof, while voltage almost hits zero... as it charges, the current decrease and the voltage starts increasing (ohms law, E=IR...) until it reaches a point that it cant take any more charge...or minimum current, at full voltage potential.

    by increasing the voltage, the current also increases... as the capacitor and its "resistance" is still the same. i say "resistance" with quotes cus its a bit more complex than that.

    the reading you get with an AC meter, or any voltmeter, will be a good approximation, but will NOT be the TRUE voltage. the CDI isnt running at a constant 50/60hz, which is what most AC meters are designed for. they also "usually" only read out RMS voltage, which is an average voltage. the magneto doesnt produce a clean sine wave, but rather a peaked, almost square wave. RMS is only useful with sine waves, at which point its roughly 0.7 of the PEAK voltage. RMS may read, say, 700 volts, but the peaks will fry a standard 1n4009 diode, rated at 1000v...

    me, im starting to agree with your pa...bad luck, much? cus i still, personally, have not fried a CDI! (read on)

    you may be better off getting a new magneto AND cdi at the same time.

    lastly...if you got a 50 volt transformer... or even a straight 1:1 isolation transformer... you can test the cdi with that. (remembering the plug needs to be earthed to the black wire still)

    you "could" just plug it into the wall socket, but DONT.




    you may wake up DEAD.

    the current available at a power point is DANGEROUS. still, even the 240 mains here(AU), is only half the voltage that the cdi unit just wasnt designed to take a virtually unlimited current dumped into it, with a nice clean sine wave... (low impedance source)

    i found that they WILL fry after about 20 seconds running on mains AC... :p even with fat 5W resistors limiting the current...

    wanna experiment?


    i take NO credit for this circuit, but it DOES work, you can buy all the necessary components for less than $5, and or...can always find the components in old power supplies etc. you may have to do a few searches on data sheets....say, a c106d works just as well as the BT151 or similar... (the one component with no label!...its called an SCR or silicon controlled rectifier) use 1n4007 or higher if you can get them, not the 4004... and kick cap up to a 600v or even a 1kv type...

    or, if you really up to it...

    rohmell - ivan H CDI mod 2.jpg

    once again, someone elses work, not mine!

    note that it includes the "jag CDI" settings,much better value than spending $75 dollars!

    slightly more complex. i have a PCB designed for it if interested, measures about 30x50mm.... a lot of work went into getting it that small! and i also did change a few things, not an exact copy. better :)

    im currently working on an entirely different take on the idea, using absolutely nothing familiar from these two circuits bar the main capacitor... im waiting for a specific component to drop below $5 EACH, unlike the SCR which can cost 5c in bulk....
  16. Kadupult

    Kadupult New Member

    Wow!! Thanks for that diagram! Yeah I'm starting to agree with the old man as well, I'm going to go ahead and order both a new magneto and CDI in hopes I can find the problem. I guess most vendors sell the same thing? That $5 build seems like a pretty fun project though.
  17. Kadupult

    Kadupult New Member

    You guys aren't going to believe this...

    5 days ago after my post I ordered 2 new magnetos and 2 new CDIs from BB
    waited for the arrival anxiously, and it just arrived!

    So about an hour ago I received my replacement magnetos and CDIs in the mail from BB (along with a bike rack I've been looking to get) and I take the kit out of the box, first thing I see, packaging is ripped, some bags have holes in them, bike rack is bent out of shape... :shout: I'm missing the mounting and nuts and bolts for one of my CDIs. but it gets worse...

    Both CDIs are dead, won't read anything on the multi-meter (just sits at 1)
    Both magnetos are at exactly 485 ohms...

    Is this a freakin joke?! I called BB personally to ask what gives, I was informed that "all CDIs read 1 or 0 on a multimeter unless you cut off the wiring" which can't be correct.

    I'm really getting frustrated now. This HAS to be just unlucky me right?

    EDIT: I just talked to BB and they said they would email me their address for replacements to be shipped, then they just sent me a blank email. I'm about to start pulling my hair out... Where can I find a working CDI?! Am I the only one in this world with 4 dead units?
    Last edited: Jan 8, 2014
  18. HeadSmess

    HeadSmess Well-Known Member

    go back, check you earths, check everything because there is no way anyone gets more than one dead cdi in a row.

    there are only two wires. one is ground, one is "hot".

    3 with the plug wire.

    between plug and black wire... 2.2-3k, not polarity sensitive. (set to ohms)

    between plug wire and blue...diode check reads 1.2v, with red lead of multimeter on plug wire. (set to continuity/diode check)

    nothing when black lead of multimeter is on plug wire.

    between black and blue, with blue on the red lead...nothing.

    with black on the red lead... diode check reads .6v between black and blue.

    on the engine itself.

    bolt on magneto, after ensuring faces are clean and theres good contact. sandpaper. steel wool. etc.

    between black and the HEAD should get the same reading within an ohm or two, as when you just short the meter.(set to ohms)

    or theres continuity. etc etc.

    by that token, between blue and any point on the engine should give you the resistance of the magneto coil plus an ohm or two at most.
  19. Kadupult

    Kadupult New Member

    My results for both magnetos and both CDIs:

    When I turn on the multi-meter it displays 1 on default

    -------------- first CDI

    plug to black wires= 1
    plug wire to blue wire diode check with red lead on plug wire= 1
    black to blue wires, diode check with red lead in blue= .807

    -------------- second CDI

    plug to black wire= 1
    plug to blue wire diode check with red lead in plug = 1
    black to blue diode check with red lead on blue = .804

    magneto wires = 488ohms
    magneto wires = 485ohms

    the only reading I seem to be getting on BOTH CDIs is on the black to blue diode check with my red lead on the blue wire. hopefully this can help someone on here isolate the problem. I haven't tested my recently kaput CDIs with a diode check, but I can if needed.
    Last edited: Jan 8, 2014
  20. HeadSmess

    HeadSmess Well-Known Member

    well, i dunno whats going on here, but...disregard what i said with the diode checks etc on that cdi in my last post!


    well, it was good last time i used it, but today...after wiping off the cobwebs, it went back into a bike.

    one pop and it became my FIRST EVER blown CDI :)

    so its readings are definitely suss.