white wire magneto unravelled. something peculiar.

[HeadSmess]

i finally pulled apart my old magneto that burnt out, in the hope that i can rewind it.

carefully pulled the core apart. just drill...five rivets. four are alloy.

carefully remove former. note direction of winding, and start unwrapping.

turn 1066, and theres the burnt out section!

3500 in total. (i have a coil winder. cant do transformers without one!)

and here is where it gets weird.

its always been assumed that the little "earth" tap, the dodgy solder join, is a tap, with both coils common to it, the other ends going to blue and white respectively.

apparently not.

the blue wire unwraps CCW, and is joined to the white wire! then the white wire continues unwrapping CCW, in a heavier guage, terminating at the earth/solder join.

(this is a "new" old magneto, the version where they have the date...5.11.2011 i think mine was... nicely made ones)


this means that...


you should be able to wire the cdi with blue/blue, black/white! with killswitch shorting these as per normal(ish)!

then the "earth" or frame is "isolated" and white is then available for power in a single wire "chassis is earth" setup! ie, a direct replacement for a dynamo. (its technically what that winding is. a dynamo.)

WITHOUT BURNING OUT OR DYING OR LOADING THE SYSTEM UNDULY.


shame i dont have a working magneto to test this theory out on.
if they are ALL wired like this, then it explains exactly why they burn out with monotonous regularity!!!!!

because black is earth, or so we think, we normally wire this to the cdi, so there is the power coils voltage ADDED to the CDI voltage.

shorting the white wire to earth should still keep the cdi operating, but the shorted power coil overloads the system, therefore acting as a killswitch. that was always a sure fire magneto killer.

adding any load to the power winding affects the trigger voltage/current to the cdi, as its reference is "floating" in regards to earth. affects the timing AND charge voltage/current. too much load and theres no charge voltage left over.

im also thinking this extra voltage is changing the advance curve? especially if the power winding is for some reason backwards, which would subtract from the trigger voltage...

the newer black/blue only coils do run better...

theres also back EMF to consider. collapsing magnetic fields and all that.

anyways. im going to rewind this thing with two coils side by side, with a definite ground tap and seperate cdi wires. use slightly heavier wire on the ignition. and double turns on power for 12v!

what is weird is i did have a magneto that powered an LED light fine... not this one. obviously there are some variations in winding!


im pretty sure that on the real old ones there was a fine wire soldered externally with the heavy one...

i shall be pulling more of these apart now

 

[HeadSmess]

and, as a side thought...maybe theres a way to use that winding with a variable shunt resistor to affect the advance curves?


hmmmmmmms?

 

[Deleted member 12676]

wrong. the CDI needs to be grounded because that ground wire is the return path for electricity going thru the spark plug.
and if you used the proposed wiring arrangement the voltage going to the CDI will be less and therefore the spark strength will be reduced, if not outright killed.
the stock CDI barely produces a spark as it is now.

 

[HeadSmess]

true that.

yes, the coils earth is internally connected via the black wire. so there does have to be some type of earth to the engine. hmmmm.

i forgot about that! spark return...

so no, cant run black/white.

well, you could, cus the power will still go through from the earth point, to the white wire... which is in series.

but then whatever voltage travels through that white winding will affect whats on the blue and spark energy itself as its the plug earth and and and and OMG!!!!!!


no wonder the stupid things burn out!!!!!

worst design ever.

you hit the killswitch, short blue to black as per standard. the high current low voltage of the white wire winding has to travel through the cap charging coil... which is delicate and not made for high currents. it fuses. at turn 1066

short a coil, it produces max current at min voltage. it also produces a back EMF proportional to the rate of magnetic flux change. that back EMF creates its own magnetic field working in reverse to the applied magnetic field (the rotor magnet). so it loads the shaft down, just like a generator on full power. (get a lil DC motor and short the wires then try spinning the shaft. makes a neat brake ).

then when you remove the short, the huge current has produced a huge magnetic field that suddenly collapses. back EMF is proportional to rate of change, so the voltage spikes up...

an induction coil. the primary winding of a ignition coil.

the things are inherently self destructive as designed!! because they have the primary winding...the heavy guage, and the secondary is the multi turn fine wire. oh wow.

this is getting deeper now ive slept on it...

the new versions dont have this heavy winding and seem to last. the heavy winding has more current capability therefore the induced voltages are higher, and yes... turns the whole thing into some kind of induction coil!

possibly a relic of the time when they were on points..possibly.

also, at the burnt section in the coil there was a cavity in the potting shellac! weakest point, isnt conducting heat to the rest of the coil as well maybe? theres always gunna be a bubble but maybe the reason some die before others due to quality of potting, and number of cavities.. just a thought.


i recall early ones would turn black?

if these guys were electricians, the house and the powerstation woulda burnt down years ago!


so, once again. im just gunna wind new coils

cant change the number of turns on the magneto side of things either.


though no. they dont operate on a trigger voltage, do they? they work on a zero crossing, yes? the voltage is simply for charging the cap... so a few more turns...will overload the stock cap possibly... cool magic grey smoke


whatever. it was interesting to strip one and see how it is actually wound.

 

[Fabian]

I am enjoying this thread and enjoying the read

 

[Deleted member 12676]

making a new one with more turns increases the voltage to the cap which ultimately translates to more spark, but since the analog circuit is voltage triggered you wind up with more spark advance for the whole curve. You can adjust for that with the Jaguar CDI though.

 

[HeadSmess]

everything ive read says the cdi triggers off zero crossing point...





in which case, extra turns will mean more spark energy.

if it is voltage triggered, well... yup. advance is all over the shop.

one one swing it charges the cap via the first diode.

on reverse swing voltage travels through second diode, triggering vs1 at voltage set by trimpot... cap discharges to earth. primary of coil being in series, generates voltage boost in secondary. its sorta voltage and zero cross triggered...

very basic circuit advance? huh?

fiddle with R1 and C1 a bit.
add another resistor or two in strategic locations...

change thyristor for summink else

as it is, ive never played with these cdi circuits.

you have.

 

[Fabian]

If the circuit design was to incorporate 5 trim pots to adjust low rpm (1,600 - 2,300), mid-low rpm (2,400 - 3,100), mid-high rpm (3,200 - 3,900), high rpm (4,000 - 4,900) and upper high rpm (5,000 - 6,500) what would the circuit diagram look like?

Could you draw it up.

 

[HeadSmess]

who knows? unneccesarily complicated

id start by just making circuit as described and play with the trimpot setting that gets best power... $5 of components if that... just make a 30:1 coil on a ferrite core. that can be experimented with too. as can charge cap value.

as im probably going to do. wind a 5000 turn charge coil, use a 2kv cap, blah blah blah. trimmers there to fine tune

if i do anything at all. could i be bothered?

ill explain it better.

on a positive swing, current flows in from top, through top diode to capacitor, and is stored. cant flow back through the diode, and thyristor is switched off.
on negative swing, current flows in through bottom wire, is forced through 33ohm resistor and trimpot. voltage developed here triggers thyristor at 7volts i think the bt151 is.

, when triggered, the cap is discharged through thyristor, through the 1n5407 rather than through the resistors, to earth. coil cops discharge currrent and boosts voltage.

as shown, its completely unregulated, and exact advance curve is set by component tolerances.
the trigger point, or initial advance, set by trimpot and rotor/magneto placement.

the charge voltage increases as the rpm increases. therfore the trigger voltage is reached earlier, and advances.

requires regulation, fixed voltages to make any definite advance curve changes.


i dont do cdi! grrrrr. ask the expert.

i play with tesla coils and other things. not cdi!

but i am thinking of finding out what happens if i hook one of these up to mains. could make a great jacobs ladder


my chainsaw seems to work fine at everything from idle to 12 grand with a "fixed" 12 degree advance...

 

[Fabian]

who knows? unneccesarily complicated

I like unnecessarily complicated (because i get to have a lot of control over the ignition curve) and i need an avant-garde circuit schematic