I need a review of my circuit. Polarities specifically

[Patrick]

Can others review this schematic for polarities of diodes/transistors and the transformer.
The transformer is a three terminal auto-transformer for truck ignition. But I dont know where to put the return HV.
From the primary they both ohm out to 8k Ohms to HV terminal but 0.4 to each other. So I don't know where to ground out the HV sparks too.

[Rafft]

3 terminal means it has one 'Gnd' only. The 'primary' is a tapped connection of the secondary (high tension) output.

 Power is applied on the primary coil, power removed and the collapsing magnetic field creates the hv output.

[Patrick]

I meant with regard to the TVS and the position where the primary and secondary meet. I think its grounded at 2 and 3.

[davekni]

Can others review this schematic for polarities of diodes/transistors and the transformer.
The transformer is a three terminal auto-transformer for truck ignition. But I dont know where to put the return HV.
From the primary they both ohm out to 8k Ohms to HV terminal but 0.4 to each other. So I don't know where to ground out the HV sparks too.

I think its grounded at 2 and 3.

Internal connection could be 2 to 3 or 2 to 4.  Doesn't make very much difference in how you use the coil.  Either way, just return sparks to +15V, or to GND presuming good low-inductance connection of bypass capacitors.

Schematic questions depend on your unspecified goals.  Is this for occasional single sparks when a button is pressed?  Or rapid-fire?  Best way to figure out your own answers is with analog simulation (LTSpice is my favorite, but there are other free tools and on-line simulators available).

D4 is likely to have much lower avalanche energy than IRF1324, so can be left out.

[Patrick]

Schematic questions depend on your unspecified goals.  Is this for occasional single sparks when a button is pressed?  Or rapid-fire?  Best way to figure out your own answers is with analog simulation (LTSpice is my favorite, but there are other free tools and on-line simulators available).

D4 is likely to have much lower avalanche energy than IRF1324, so can be left out.

O.K. so here is the code and new schematics, it is intended to function as follows: 50 pulses on, 2 seconds off. Ive also added a capacitor across the D-S of the Mosfet is that in a good idea ?

[Rafft]

Try it this way.

Refer to webpage for x-former connections
http://mgaguru.com/mgtech/ignition/ig104.htm

Never done any like this(mosfet) & ignition coil, but its the only way possible to drive an ign coil electronically. im also thinking the 'return' or 'gnd' side for hv output is battery+(and not circuit gnd). Just try it. 1N4007 or FR across mosfet. Iron core is supposed to be low freq. Like 50 or 60Hz. Maybe max 400Hz

[Mike]

I'm not sure if Q1 is connected the way you want. You're using an emitter follower configuration so you will only apply about 4.4V to the gate of your MOSFET. I suspect you wanted to apply 12V so you should be using a common emitter configuration. This will have the effect of inverting your logic (putting D4 high will turn the MOSFET off) but this is easily corrected in your software.

I also noted the LED in your trigger button doesn't have a current limiting resistor. Some switches include one, but not all, so just be wary. Your trigger won't work at all if there's no included resistor.

[SteveN87]

The turns ratio of an ignition coil is more like 100:1 than 1000:1.

As connected, D5 will short out the back EMF (great for relay drive circuits, but not so good for this), meaning you'll get about 200V of spark voltage (you would be more likey to get a spark from the forward voltage - about 1500V). Probably just remove D5.

[Patrick]

As connected, D5 will short out the back EMF (great for relay drive circuits, but not so good for this), meaning you'll get about 200V of spark voltage (you would be more likey to get a spark from the forward voltage - about 1500V). Probably just remove D5.

Yes was thinking the same Ill remove it and try. This circuit just makes me nervous with a delicate microcontroller involved as this breaks one of my cardinal rules:  -Don't trust programmed silicon near HV-  I'm sure the Mosfet and Bipolar transistor will eat a harmful kickback, and I could add a opto-isolator later.

But the timing is a hell of a lot easier than 555 timers which I was limited to all though the late 90's.

[davekni]

I'm not sure if Q1 is connected the east you want. You're using an emitter follower configuration so you will only apply about 4.4V to the gate if your MOSFET. I suspect you wanted to apply 12V so you should be using a common emitter configuration. This will have the effect of inverting your logic (putting D4 high will turn the MOSFET off) but this is really corrected in your software.

I see 12V going into the processor block.  Had presumed this output was 12V.  If 5V, then this is accurate.  IRF1324 is spec'ed at 10Vgs.

The turns ratio of an ignition coil is more like 100:1 than 1000:1.

That's true for old ignition coils used with mechanical break points and some new ones with similar electronic control.  However, I've seen other new coils with >=1000:1, I think designed to produce sparks with the forward 12V across primary as well as the reverse voltage at turn-off.  These coils are driven with short pulses (much less than 7ms) of ~20A at 12V.

As connected, D5 will short out the back EMF (great for relay drive circuits, but not so good for this), meaning you'll get about 200V of spark voltage (you would be more likey to get a spark from the forward voltage - about 1500V). Probably just remove D5.

If 40kV is desired, then 1000:1 isn't enough for this 24V FET IRF1324.  Turn-off voltage will be lower than 15V on-state voltage.  Much of stored magnetic energy will be dissipated in FET avalanche breakdown after turn-off presuming D5 is removed.  If D5 is kept, then energy is dissipated in D5 and in coil primary resistance.  Either way, circuit will be inefficient unless most energy is transferred to secondary during forward FET on-time.

Depends on primary current and inductance whether 20ms is long enough for current to decay.

[klugesmith]

Yeah, what Dave said about voltage margin.
Consider the "magnetic field collapse" in classical ignition circuit or flyback converter.
How fast does the magnetic field collapse when switch is opened?
If switch turns off instantly, and isn't shunted by a capacitor, then voltage goes to whatever it takes for something to break down in primary or secondary circuit.  THe stored energy has to go somewhere!

Commonly it's the intended spark gap or unintended D-S bypass in MOSFET. (Secondary/primary voltage ratio remains N:1.)
That voltage then determines the rate at which magnetic flux drops to zero. 

[Patrick]

Ok Ill continue to post some changes for you guys to see. It sparks more than 23mm every time and if i tune it better i think it will go way past that. But with the IRF540's 0.85 resistance its flaming hot even at 11.5 - 12 volts in about 20 seconds. I dug around and found some IGBTs that are rated for CE 1.7 Volts at 10 or so amps. Ill cut a bigger heatsink and put in the IGBT in a few hours.... need sleep ...... ZZZZzzzzZzzzzZZZzzz...

[petespaco]

Just a couple of comments, questions:
   What IS the goal of this project?  That is: --- why are you making sparks?

Can you tell how much time the Mosfets are spending in linear mode?  I don't think they like long switching times.

It took me a while to realize that the "auto-transformer" you mentioned in your first post meant "automotive TYPE" of ignition coil  and NOT "autotransformer".
If you are worried about HV pulses sneaking back into your control electronics, why not simply get yourself a "Coil On Plug" or "Coil Near Plug" coil from a more modern ignition system?  They usually accept a 5 volt trigger pulse.
I have many examples of doing this and here is one link:
xW0KdiP5MJY

Here's a much later link:

/>
And, if you are a real "Gluten for Punishment", here's my  playlist for my own "Coil On Plug" monkey business:
https://www.youtube.com/playlist?list=PLVxWen9M87dAFoZqyj3iy_fyBfkWuESRC

Pete Stanaitis
---------------

[Patrick]

   What IS the goal of this project?  That is: --- why are you making sparks?

I need to ignite Methanol, Ether or Gasoline for a pulse jet.

[RoamingD]

   What IS the goal of this project?  That is: --- why are you making sparks?

I need to ignite Methanol, Ether or Gasoline for a pulse jet.

Hello, I am not an expert but what I can say is that you will have a way easier time with using a flyback transformer from an old tv ( you can program the arduino so that when it is triggered, to output a 15-20khz square wave signal to a small bjt that will drive the gate of a mosfet of your choice (preferably something like an irfp 250/260n) connected to a 8 turn primary coil wound on the ferrite core)
 For example, here is a video I found on youtube with an arduino driven flyback transformer

I am getting 5cm arcs from my irfp260n flyback transformer with a 12V power supply and it doesnt get hot with a good heatsink