SEM ignition that fitted KTM400LC4 – 1996 to 2002

Flywheel and stator on right side of the bike.

right side


red charging coil = 200Vac
green pickup coil = +/-6Vac
black frame ground
yellow lighting coil
blue lighting coil

Cliquer sur les fils

yellow= 12Vac 12Vac green= pickup green= pickup red= charging coil red= charging coil red= charging coil

Click on wire to visualize the signal available on this wire, when the engine idles…


Inside the stator


Thanks to Zoli Bihari


Resistances map.

OHMs Red green black yellow blue
red 1777 1754
green 1777 23
black 1754 23
yellow 0,2
blue 0,2

Until 1998 pickup coil (green – black) was 23ohm. After it’s 165ohm.

Here is how STATOR looks like when measuring connectivity , resistance and capacitance between wires:

Ignition coil

It’s not a simple coil! There are some electronics inside…

The “black box” is part of the coil:

red charging coil = 200Vac
green pickup coil = +/-6Vac
black frame ground
orange handelbar kill switch

Here is how COIL/CDI looks like when measuring connectivity , resistance and capacitance between wires:


Orange “BLACK BOX” from a KTM 620LC4 – 1997




To replace a TM14-04, use an ACCDI v2.6+ plus an standalone ignition coil rated at least for 500cc.




  • Pickup and Spark Signal at 1460RPM.
  • Pickup and Spark Signal at 2880RPM.
  • Pickup and Charge voltage at 1770RPM.
  • Pickup and Charge voltage at 2370RPM.



LC4 kickstart Ignition coil with SEM before 1999 SEM TM14-04 TM14-07

+ pole – pole Measure
Coil side
red black 47k
red green 47k
red orange 0ohm
black ignition lead 2,0Kohm / more or less 0.3K
Stator side
black red 1,7kohm
black green 23 ohm more or less 5 ohm
red green 1,7kohm
yellow+yellow blue 0,2ohm

LC4 kickstart Ignition coil with SEM (ignition coil) after 1998 SEM TM14-05

+ pole – pole Measure
Coil side
red black 47k
red green 47k
red orange 0ohm
black ignition lead 2,0Kohm / more or less 0.3K
Stator side
black red 1,7kohm
black green 165 ohm more or less 20 ohm
red green 1,7kohm
yellow yellow 1ohm

LC4 electric start with KOKUSAN
Impulse transmitter coil red green 100ohm more or less 20%
Stator black / red red / white 12,7ohm more or less 20%
loading coil mass / yellow 0,65ohm more or less 20%
loading coil white / yellow 0,16ohm more or less 20%
LC4 kickstart with KOKUSAN super comper 1999-2003
Impulse transmitter coil white green 60 -120ohm
stator red / black-yellow 0.45 – 0,56ohm
Stator black / red and red / black 0.45 – 0,56ohm
Ignition coil KOKUSAN
Primary blue / white – mass 0,45 – 0,56ohm
secondary blue / white – zundkabel 0.45 – 0,56ohm
TM14-04 is exactly the same as TM14-07 (same advance circuit, only the PCB has change)

74 thoughts on “AC-CDI KTM SEM”

    1. I’ve some available, cause mine GS620 SC stopped sparking instantly. You can also look for a replacement from Electrexworld but KTMdealers don’t love it. Steve from MOTOPLAT: 0044 (0)7516 350 134 repairs them for a fixed price of 185 GBP

  1. I’ve a KTM GS620 SC from 1996 with a kickstarter, it is a 4T . Who can give a suggestion or has already experience which CDI published on is usable for this engine?

  2. buenas tardes. Me podrian ayudar con una bobina stator para una motocicleta KTM LC4 620, que este nueva
    para colombia

  3. Hello found the schematic to be a little outdated any chance you can help source more modern components?
    I am looking for an AC CDI for my KTM 620 (I think the ignition timing is either 36 or 38) will the AC CDI product you sell work for me?

    1. Hello,
      I would not say “outdated”, it’s not about Fashion !
      This schematic is the genuine one from KTM using techno from the 2000’s
      A processor is needed if you want to do fancy things but for triggering a SCR at 5 volts a diode do the job perfectly and in reliable manner.
      Furthermore there is no battery on most of offroad KTM,Maico,Husky as µproc need a stable power…

      AFAIK Your KTM620 can use a SEM ignition (AC-CDI) or Kokusan (DC-CDI)

  4. Hi there, great page thanks for this info.

    I have a 1996 KTM 620 with an SEM stator, getting no spark. After testing the resistance across the stator wires (all seem ok) i note that the black lead has continuity with the frame of the stator itself, and therefore the bike. Is that normal ?

    i don’t understand how that could be given that the kill switch effectively connects the black lead of the coil (through the orange) to earth to kill the motor.

    any ideas would be appreciated.

    1. Hi, thanks!
      Yes it’s normal, CDI is grounded to the frame through the black wire.
      No, the orange wire kill the engine by connecting the RED wire to ground, (not the black one)
      Disconnect the orange wire to check if you don’t have a permanent kill….

      1. ok cool sorry to be obtuse, but the cdi being grounded to the frame via the black means that the *stator being grounded to the frame via it’s ‘own’ metal body is normal. gonna try disconnecting the orange wire tonite.


          1. ok i have a spark but it seems to my amateur eye that the spark is weak and / or intermittent. i am able to hold the sparking end of the plug against the frame and am only slightly uncomfortable with the shock, from memory it should give me a whack (i didn’t start by holding it in my fingers!). Up until now i have had intermittent probs including problems hot-starting, which i understand can point to a bad stator. Coil resistances are correct, plug lead and plug are ok. Gonna bite the bullet and replace the stator. I’ll let you know how it goes.


  5. Hi again, sorry to abuse your hospitality but you clearly know more about this than me.. I read your message below late and when ahead and ordered a stator. it arrived and i’ve installed it, in my opinion the spark is better and importantly consistent when the plug is outside of the bike. however, i have a serious ‘kickback’ problem, which i guess is caused by bad (too advanced?) timing. The new stator does not have a timing mark on it, and the old one didn’t either so i have no idea how to time it ! even on the most ‘retarded’ setting it kicks back (i cant go back any further), never used to do that with the old stator. is that further indication that the cdi is faulty? i don’t have an oscilloscope btw. i appreciate any advice you might have,.

    1. It’s a SEM stator?
      I think that the 3 bolts are not aligned so it’s not possible to fix it the wrong way… Right?
      Have you at least compared your measurements on the CDI with the resistance map above ? Everything match?
      Without any measurements neither oscilloscope traces done before and after replacement it’s just a shot in the dark !

      1. sorry i should have been more explicit, i have measured the resistances of the cdi and they match your table above (SEM before 1999). You can potentially mount it 3 ways, by rotating it on the 3 holes, i have mounted it the way it appears in all the photos on the net (and the way i found the old one). after that there is a degree of movement to adjust the timing, because the holes are not circular. gonna get back out there tonite after finding some detailed pix on the net of the position of the marker, hoping its the same on mine. problem is this game is breaking my ankle !

          1. Hi again, sadly that’s not the case (i’m pretty sure), lookup in the pic on this page and you see that the inspection window is not correlated with the ‘poles’ of the stator, its kinda 5 o’clock if the pickup is noon. after playing in the garage there is only one way it can go, the way in your photo, and i’m not able to retard it enough for it to work… sigh.

          2. Well we can assume it’s fit the good way then!
            Seem you were right and the stator failed but the new one COULD be different and give too much voltage and the CDI trigger sooner…
            Try to add a 1K adjustable resistor in series with the GREEN/pickup input wire.

  6. Hi again, in my frustration i went ahead and bought a new cdi to go with the new stator.. ST5500L and CD5004. my thinking was that replacing the 2 together should eliminate any incompatibilities.. wrong. the kickback problem is still there. if i switch back to my old stator, the kickback problem goes away (with either cdi!), it can sometimes ‘just’ idle for ~10 s but then dies, the reason why i wanted to replace it in the first place. think i’m gonna try your idea of putting a 1k resistor on the pickup circuit to more closely emulate the original stator.

      1. hi again, i put an adjustable resister in series with the green pickup line. tried kicking it with the plug outside of the bike to see at what point the spark dies, for me i only needed to add about 30 ohms before it stopped sparking. wouldn’t start with the resister in place though. remove the resistor and it goes back to the kick-back behavior. i’m wondering if the company that supplied the stator mixed models, gonna ask them to send a replacement.

          1. Thierry: YOU ARE THE MAN !!!!! I put 3 diodes in parallel with each other, in series between the cdi and stator and… VROOM ! no kickback, everything works ! would you mind explaining what this means ? please send me an email so i can send you a couple of beers :)))

          2. Cool ! Thanks to consider buying me a beer! There is a Paypal Donate button on the left side of this page 😉
            Definitely this adaptable stator is somewhat different…
            follow-up on PM

  7. Hi
    I have a KTM EXC 250 1998 with no spark.
    I have checked SEM stator resistance values:
    Black-Red- 1.7K Ohm
    Black-Green- 168 Ohm
    Green-Red- 1.9K Ohm (instead of 1.7K Ohm)
    Yellow-Yellow-0.4 Ohm (instead of 1 Ohm)

    Should I get a new stator or the values are OK?

    1. Hi,
      Those values are OK.
      yellow / yellow is certainly 0,2ohm like models before ’99
      Anyway yellow/yellow are the 12Vac coil and you don’t have issue with +12v Isn’t?

      1. Long time, but I haven’t seen your reply and unfortunately the bike still doesn’t have a spark.
        I never checked the 12V because the bike doesn’t have lights.

        If the values are OK, should I order a CDI unit? Or the stator might still be the problem?

        1. No I think the stator is fine. if you want confirmation, remove the sparkplug, rotate the crankshaft with a drill and measure the outcoming voltage.
          ACCDI v26+ can do the job with an external standalone ign. coil

          1. Thank you very much for your help.
            How much voltage should I expect from charging coil and pickup coil ?

            And another question- can I test the existing CDI using the 2 yellow wires and a 12-220 V transformer in order to replace the charging coil? Do I need to use a rectifier? If it can work, how exactly should I connect all parts?

          2. IDK what YOUR multimeter will say!
            Only True-RMS multimeters are able to accurately measure non-sinusoidal ac waveforms with irregular patterns, squares waves…

            Look closely to the scope trace above: pkp: 2v/div , charging: 100v/division

            You want to use the 12vac from the lights coils ?
            Yes it should work more or less… Transformers are 50Hz and the AC signal is way more !
            Or use 2 transformers: First a 220-12 connected on the 220vac main power, and a second 24-220 (in reverse mode)
            So you should get 110Vac at the output.
            Use a rectifier?
            Only if you want to destroy your ACCDI 🙂

          3. Yes, I want the lights coil to replace the charging coil.
            Is it the correct way to connect ?:
            Connect 1 yellow wire of the stator to each side of the 12v side of transformer.
            Connect the 220v side of transformer to- bike frame, CDI red wire.

  8. Hi Thierry,
    I have a problem with sem tm 14-04 , all the measuremens are the same that sem 14-05?
    The only difference i have found its coil green black , about 1.5Mohm (almost open)instead of 47Kohm
    Is this ok for 14-04 or may i have to change for a new one?
    Thanks in advance

    1. Hi,
      Yes except the pickup values are identical.
      1.5Mohm is clearly out of spec measurement !! I think it’s faulty
      Have someone to lend you one or build the ACCDI analog v2.4 is you know about electronic.

  9. thanks, Thierry
    sorry I said wrong colour:
    I said ignition coil green black , but really it´s ignition coil red green.( about 1.5Mohm (almost open)instead of 47Kohm),
    therefore the ignition coil may be damaged.
    Anyway my stator readings are ok, but the stator pickup it´s 165ohm
    the bike it´s Aprilia Red Rose 125 1990 ,therefore , I assume that someone changed the stator for a new one ,
    and leave the 14-04 ignition coil.
    With this stator reading (165ohm) may I install 14-05 ?
    or it will work with 14-04/14-07 ?

    1. Hard to know…
      I think it should work with TM14-04 but a 165ohm pickup provide more voltage and “could” harm a 14-04 sooner or latter…
      Better is to go for a 14-05 to adapt stator and CDI

    1. Honestly I don’t care when a new one for a 250cc cost only 10$

      Edit: The primary handle 200V and 10000V is the minimal value to get a spark at the secondary, so the ratio must be at least 1/50 until 1/100.
      Wire are thinner then that: around 200turns of 0.8mm and 20000 x 0.05mm

  10. HI Thierry is the stator 3 phase or single phase? There are 2 yellow wires and 1 light blue wire coming out of the stator for lighting. My bike is wired with the two yellow wired together then to the lighting and the blue to the chassis ground. My bike reads 12vac and the lighting has flickering at low RPM and the LED lights I added are very dim until I increase RPM. I have no Battery. Thanks for your continued support to this post

    1. Hi,
      My fault! it’s indeed 2yellow and 1blue (corrected above).
      IMHO it’s a one phase. If you haven’t I suggest to replace the battery by a capacitor that will act as a current storage during the 80% of time when stator doesn’t give anything. (pulses of 15v during 20% of time)
      This 47000uF/25v/18€ capacitor SHOULD fit…
      If there was no battery and just AC, you MUST add a 5A bridge rectifier before the cap.

      1. Thanks for the suggestion. The one I see in the parts fiche is 10,000 Mf, on the actual item is written 10,000uf. Which is confusing. Is it okay to oversize the capacitor in this situation?

        1. what is confusing for me is the way you write it down!
          I strongly doubt that a official datasheet says: 10,000 Mf with a comma and a capital M !!
          LOL it’s simply the International Metric System 10mF = 10000uf
          yes it’s OK.

  11. Hi Thierry, my compliments for all the explanation you are doing.

    question: do you know anybody that installed any of your Ac-CDI in a Aprilia 650 (~BMW 650Funfuro). the old version earlier than 1995? The originals are difficult to get and old, so I am looking for a replacement. Apparently these old one worked with one trigger coil and one trigger from the generator..?

    1. Hi,
      Thanks. One person was interested to build one for Aprilia pegaso 650 GA 1994 (with one 240ohm pickup) but I don’t know if he followed up….
      What is the reference of the box? I believe that
      Nippon Denso 070000-0780 / 1992-1994 / 9wires are ACCDI
      Nippon Denso 070000-0790 / 1995-1997? / 6 wires are DCCDI

      1. Indeed, it is Nippon Densi 07000-2170 / 1992-94/ I think 9 wires and a 240 ohm pickup. And that is what puzzled me. Two different guys told me that the CDI has one external trigger and one of the AC coils is also a trigger. I have schematic of the stuff but am not sure how to compare these to yours.. your opinion is very much appreciated!

        1. You can send me the schematic by PM. My email in on top right link “contact”
          Is it a ACCDI too?
          CDI can be triggered by external pickup or by the stator, but I never seen both at the same time! That is no reason to do so !

  12. thierry, schematic is more connection setup to CDI and alternator not a schematic of the ignition (helas) you can find it here. Obviously it is a AC-CDI.

    A guy from told me the dual trigger story. It seems to be true as I still have a spark on the highest delay without using the separate trigger coil (and using another working CDI.)

    all your idea are welcome.

    1. It’s a wiring diagram not a schematic. it is here.
      Well perhaps the external pickup has been added later on production to ease the kick start…?
      Yamaha XT600 also had 2 pickup for low and high revs.
      If and only if the internal pickup for hi revs puts out ONE pulse per rev, ACCDI v7.9 can be use. low(aka12°) attached to the external pkp, Hi(36°) to the internal one.

      1. thanks, maybe I will do that and use only one pickup. With the successor of the bike >1995, they skipped the 2 for just 1 trigger (and used battery driven CDI). thanks for the input

        1. Works with a single sensor! (I bought the PCb already in March for another project..:)). so, naturally I have to change the position of the magnet. (essentialy it is then the setup of the 1995+ Aprilia/BMW).

          There is one other problem now. My impression is that the magnet is partly demagnitized or so. The signal is weak. did you ever encounter that, and what did you do? use another trigger coil (more windings, more ohms?)? or did you glue another magnet (seems a little tricky for underway..)

          in any case thanks a lot for your help…

          1. Great. I suppose you use the external pickup which is for LOW RPM, so you need to move it?
            With some luck you can try the EXTRA-ADVANCE feature

            No I haven’t had. Beware ! Space between rotor tab and pickup is EXTREMELY sensitive. If the pickup is moved away from the tab for a tiny mm the signal gets weak.
            A hall effect sensor is easy to glue, some guy here have done it. The MAJOR trick is to fit in very very tight, then with glue sealing it seem to carry on the heat well.

          2. Hi Thierry, bought your PCB 7.9 (Printed Circuit Board AC-CDI 16F628 v7.9 ) in march last year. fiddled around with it quite much and finally build in a Pegaso (here above). now I d like to do that again, but V7.9 is not there anymore in the shop? do you still have one, or can I find PCB anywhere? Much obliged..,.

  13. Have you ever worked on k10 with sem tm11-10 cdi? It’s 94 husky te350. Seems to have an additional wire for charge coil (red wire,blue wire), trying to figure out if I can wire this to a stator that has just red black green ignition wires. I have the stator unpotted. Good work on the Kim unit

    1. No I haven’t seen a TM11-10.
      If they kept the same color by function, then:
      Black: GND Red: charging coil Orange: Kill sw Green: pickup
      and there is only one extra wire: Blue.
      I don’t know what it’s used for…. Any ohm measurement between blue and other wires?
      I would probably connect the TM11-10 to the same colors on the stator and live the Blue wire disconnected.

      1. It worked. Started right up. The blue wire is connected to the each charging coil between them on the outer edge with the red connected at the last one. My lights still don’t work. The whole reason for the stator swap. I get 5vdc at idle and with throttle it drops off

  14. man, can i reproduce the stator voltage from another source? like a car converter 12v-220vac
    and use the pickup coil from the stator
    i start an scooter with that method, but idk if its working on ktm lc4 540 sem ignition , tm 14 04

  15. we have a problem with a replacement for the tm 14-05 i have a cd5004 from hpi ( and i have a problem with it and the same with my original tm 14-05. the problem is that i have infinte ohm between black and red , and between green and red do you maybe have an idea what the problem could be the replacement is new 1 time in use and have this issue could it be a problem with another part

    MfG Tobias Hagen

    Ps. Sry for my bad englisch i am not a native englisch speaker

  16. Hi Thierry, bought your PCB 7.9 (Printed Circuit Board AC-CDI 16F628 v7.9 ) in march last year. fiddled around with it quite much and finally build in a Pegaso (here above). now I d like to do that again, but V7.9 is not there anymore in the shop? do you still have one, or can I find PCB anywhere? Much obliged..,.

  17. Hi Thierry,
    I have a KTM 300 95. It doesn’t spark.

    I have checked SEM stator resistance values:
    Black-Red- 1.7K Ohm
    Black-Green- 162 Ohm
    Green-Red- 1.9K Ohm
    Yellow-Yellow-0.5 Ohm

    I have checked the TM 14-07
    Black-Red- 47K Ohm
    Red-Orange- 0.3 Ohm
    Green-Red- 300K Ohm
    Black-Ignition Lead -Open Circuit

    The coil appears to be broken. Can I install a TM14-06?


    1. Hi Ricardo,

      Yes I guess.
      TM14-06 & TM14-07 have the same advance curve
      TM14-04 has a different advance curve
      TM14-04 TM14-06 TM14-07 share the same internal coil

  18. Hello,
    I have ktm 250 1994 with SEM Ignition.
    My flywheel magnet rotor broke. And there are no magnets left. Where i can find a new one ?

  19. Hello! Could somebody please send a full description or all important details about the differences between TM14-xx variants?
    I’m interested in the difference between the units. (Especially Tm14-01 and Tm14-04).

Leave a Reply

Your email address will not be published. Required fields are marked *