Fiddy wrote ...


As Sigurthr said, the 10k resistor discharges the gates capacitance, without it, after turning the power off and disconnecting the Gate from your driver, the Source/Drain will still be conducting so if power is still across the Source/Drain it will allow current flow even tho there is nothing controlling the gate pin, this usually causes max currents to flow and will destroy the FET.

Ash Small wrote ...

That is very interesting. I didn't realise that the MSD coil fired multiple times. I'm working on a multi-spark system in the thread here: (proposed circuit diagram on page 5), butv I'm waiting for MOSFET's with sub 10nS switching times to become more affordable (I need twenty or so for each coil, and last time I checked they were £40 each). Part of the reason I'm building the 556 timer is to learn more about the whole concept.

The primary resistance will be much lower if it's a multi-spark coil. (I did wonder when I saw your video how you were managing to drive an ign coil at what looks like quite a high frequency for ign coils)

Maybe I'll have another look at the MSD coil. I did look it up earlier in my other thread after JohnF mentioned it, I think, but didn't realise it was a multi-spark system. As I understand it these multi-spark systems use capacitamce, and don't fire until the voltage rises to the breakdown voltage, after several pulses, then continue to supply more current at lower voltage during the discharge. I think I posted some waveforms earlier in the other thread.

There have been several multi-spark systems in the past, which claim to fire several pulses, but as the technology of MOSFET's etc gets faster, the ability to supply more pulses per 'ignition event' is increasing, although 'state of the art' components, as used in pulse generators of the so called 'voltage adder' type, discussed in a recent thread of Patrick's on planar transformers are still very expensive.

EDIT: Just noticed this bit:


I'll have to give this some thought, too.

EDIT 2: If power is shut off at Vcc, there is no power across source/drain, so the problem doesn't occur. It did occur to me to add a capacitor and diode, so that it can shut down, without the risk of further triggering, but I don't think this is necessary, as the circuit stands, but I've had a drink. I'll look again after coffee in the morning. As it currently stands, I don't think this problem will occur, though.

Uspring wrote ...

The effect of the air gap is mostly to reduce primary inductance. For a given input voltage and on time this implies larger primary currents, which will store more energy in the primary. The same effect can be achieved without a gap, when you add a small inductance parallel to the primary. Most of the energy will then be stored there.

Dr. Slack wrote ...


....you may only need a tiny H field to get a ferrite core up to a max B of 0.2T, but a large H field is needed to get a lump of air to the same max B. As energy is ..... B x H, the air stores much more energy getting it to the same B field.

Now core volt-seconds are related to the core area and maxB, ...... The core has a maxB and an area. The air-gap, assuming it's short and a large area defined by a gap in the core, has exactly the same area as the core, and has a Bmax defined by the core as well, even though the air can of course support a higher B field. So it doesn't matter whether there is an airgap or not, the volt-seconds remains (to first order) the same.

Dr. Slack wrote ...

Amp.seconds has no place in transformer design or understanding. So you have been mis-informed.

Coulombs are useful if you are trying to charge a battery