Even if its a HEI coil and not a common grounded type ignition coil?

Yes.
Any coil arrangement will generate a voltage spike that you need to limit to protect the transistor.

 

What's the purpose of Q1?
Just use a resistor.

Why do you have 3 series diodes in parallel with 1 diode from the MOSFET drain to ground?
The 1N4007 is redundant.

What's the purpose of the two parallel diodes going from the limiting resistor to the MOSFET drain?

 

Arent the ignition coils made to work on 50% duty?

They don't care about duty-cycle as long as the voltage is applied long enough for the current to reach the desired value.

The 4Ω resistor in the one shot 555 circuit is there to limit the peak current?

What 4Ω resistor?

You didn't answer what Q1 is for. (?)

 

I meant the 4Ω resistor in your one shot timer circuit.

Sorry I was looking at the timer and not the output.
Yes, that's to limit the current to 3A.

Q1 is there because I didnt change the timer circuit with another which can vary duty cycle.

I don't understand.
Do you want to vary the duty-cycle or not?
In either case, Q1 is not needed (or wanted).

 

Here's a circuit that also allow adjustment of the duty-cycle.

 

Whats an easy way to disable output once the supply reaches 9.8v?

Is that to disable it if it's battery powered and the battery becomes discharged?
Do you want to just disable the output, or cutoff power to the whole circuit?

 

Here's the 555 circuit with a voltage cutoff circuit
It uses the TL431 programmable shunt regulator as a comparator with a built-in accurate voltage reference, to pull the 555 RST input low, which stops the output oscillation when the battery drops below the set voltage.

When the TL431 Ref voltage drops below 2.5V, as determined by the U4/R3 voltage divider ratio, it stops conducting and its cathode voltage goes high, causing MOSFET M1 to turn on, pulling the RST signal low.
This also turns on the D3 LED low-voltage indicator.
R4 provides some hysteresis so the circuit doesn't oscillate at the set point.

The simulation shows the RST line (yellow trace) going low at 9.8V (red trace), which can be adjusted by pot U4.
LED D3 indicates the Low Battery condition (green trace).

 

I read TL431a is easily damaged with overvolting. Does this call for TVS clamps?

That's an odd statement. Don't believe all you read.
Most semiconductor devices are readily damaged by an over voltage so I don't see anything special about the TL431 in that regard.
The TL431 is rated at 37V max cathode to anode and it can't be readily overvolted in this circuit.
When the voltage is high, the TL431 is full on and has only a couple volts across it, with most of the voltage being dropped across R3 (see V(k), purple trace below)

What do you think would generate an overvoltage?



Edit: I just noticed a problem with your adaption of the circuit.
The RST signal voltage should go no higher than the 555 supply voltage.
So solve that, connect the LED resistor (only) to the the 555 +5V supply instead of the battery voltage.

But what's the purpose of the 5V regulator?
You can run the 555 circuit directly from the battery voltage.

 

Moved TVS diode to 2n7002 gate should be more usefull in that place.

That's overkill as any power supply transient will be small by the time it gets to the gate.

You still need to move the low-battery LED resistor to the +5V supply.

 

Does it make a difference if the switch is high or low side?

The battery switch?
Not if the ground side of the circuit is otherwise floating.

 

Only the high voltage mosfet needs a gate TVS?

I don't see why that would need a TVS on the gate. It only sees voltage from the 555 output.
You seem to be somewhat obsessed with transient voltages. Have you had a problem with them in the past?

TL431a spice files attached.

 

Wouldn't disconnecting + side of the battery case an arc without the TVS? the transformer stores energy in its large steel core.

I'm confused.
Is the transformer used to charge the battery?

The transformer can cause a transient voltage on the primary due to the primary magnetizing inductance if the primary is disconnected, but nothing significant if the secondary is disconnected, since that doesn't interrupt the magnetizing current.

Pot files attached.

 

Nothing is charging the battery. The thing will just run on a battery or 12v adapter.

Then why did you mention a transformer?

are there any specific layout things which are important to observer?

None I can think of.
The layout is pretty non-critical for that circuit.

 

The transformer I mentioned was the HEI transformer.

Okay.
But that's only connected to the output MOSFET drain.
If have the MOSFET source ground directly connected to the power supply ground with good capacitive decoupling (say 100μF electrolytic in parallel with 100nF ceramic) between the HEI supply pin and the MOSFET source pin, you shouldn't have a problem.