In place of the 1N4005

 

I've received those Schottky diodes now.

I have also decided to adjust the voltage across the coil primary rather than just have it fixed at the 12V supply and also insert a momentary switch into the primary coil circuit for activation. I'm wondering if the back EMF generated by the secondary will have an impact on the primary (as one side is common to both) and hence affect the rating of the momentary switch? Or perhaps a suitably orientated diode will offer some protection as in the attached?

 

None of those three diodes protect the MOSFET from the back emf generated when the MOSFET switches off.
A car ignition system uses a capacitor, typically 0.2uF to 0.3uF, across the points. I would suggest the same approach here. The voltage generated on the primary should be around 300V so a 400V rated MOSFET is pushing your luck.

 

None of those three diodes protect the MOSFET from the back emf generated when the MOSFET switches off.
A car ignition system uses a capacitor, typically 0.2uF to 0.3uF, across the points. I would suggest the same approach here. The voltage generated on the primary should be around 300V so a 400V rated MOSFET is pushing your luck.

I was assured that the SR5200 diode would be fast enough to bleed the back emf to ground.
Would your suggested capacitor go across the Drain and Source in parallel with the SR5200? What voltage rating should the cap be and do I not need to protect the momentary switch from any back EMF?

An IRF840 might be better rated at 500V Vds min.

Thanks

 

I was assured that the SR5200 diode would be fast enough to bleed the back emf to ground.

It will protect the MOSFET from a negative drain voltage though hte intrinic diode in there will do job.
When the MOFET turns off the back emf will drive the drain positive which will reverse bias the drain-source diode and forward bias the other two so no protextion there. You could put a reverse diode across the coil primary as is done for driving relays but that would completely kill the primary voltage and that would kill the secondary voltage. You need that large voltage across primary to get an even larger voltage on the secondary.

A capacitor on the primary side will keep the voltage within bounds and allow the circuit to work. Probably the best source for that capacitor is an auto supply store. They are pretty cheap and you will get one that is intended for that function.

 

A capacitor on the primary side will keep the voltage within bounds and allow the circuit to work. Probably the best source for that capacitor is an auto supply store. They are pretty cheap and you will get one that is intended for that function.

Ok I will hunt around for one. Do I need any protection for my primary coil switch as I would hate to have it breakdown from a primary or secondary back emf. If only the primary is involved then a switch rated as say 600V should be fine I presume.

 

Ok I will hunt around for one. Do I need any protection for my primary coil switch as I would hate to have it breakdown from a primary or secondary back emf. If only the primary is involved then a switch rated as say 600V should be fine I presume.

In an auto ignition system, the capacitor helps to protect the points so I guess it will protect the switch too.

 

Right placed plot diagrams are for R_LOAD=1GΩ, left - for R_LOAD = 1 MΩ.
ADDED: Optimal value of capacitor C1 is 10 nF, not 5.1 nF.
Attention!
1. Device must have stationary 15 mm spark gap to prevent generation HV more than 15 kV, for transistor safety.
2. Capacitor C1 cannot be connected directly to drain without transistor damage by discharge pulses. Instead used snubber D1, C1 R2.
3. In device should be excluded positive pulses on gate longer then 2.5 ms or DC, else transistor will damaged.

_______
http://www.myelectricengine.com/everythingelse/ignition/measurements/measurements.html
https://mgaguru.com/mgtech/ignition/ig104.htm

 

Thanks for that. I hope this is off the shelf or previously done work and that you didn’t go to this trouble just for my circuit.

I’m not clear if this shows that the two diodes will protect the FET when a spark does not occur. As per previous advice I’m going to put an ignition coil condenser across the Drain-Source to add protection to it. They are typically 0.3uF 600V units.

Lots to consider.

 

The graphs for pulse of the motor ignition coil are examples pulse charecterization but in the entire circuit over time effeciency
we sense the components for heat over time this is called Faradaic effieciency in this type circuit.
The Charecterization of the circuit components allows choice to be made is useful with regard to Faraday effeciency
which is not controversial or secret in any way it is just good electro physics. Because an electrical engineering application
might require specialized study many of these extensions are difficult to discuss in an open forum. The thermal properties of a circuit
would eventually need more descriptions then the topic would run into moderation.

 

The graphs for pulse of the motor are examples pulse charecterization also the circuit over time effeciency
we sense the components for heat over time this is called Faradaic effieciency. The Charecterization of the circuit components is useful with regard to Faraday effeciency which is not controversial or secret in any way it is just good electro physics.

I wasn’t aware of any motor in the system - just a trigger system for an auto ignition coil.

 

Hello,

Rectifier diodes like 1N5400 and related are meant to be used at line frequencies like 50Hz or 60Hz. Using them at higher frequencies like even 1kHz means they will cause inefficiencies. You need to use high speed diodes the Schottky types are one of those kinds.

What is that 1N5400 doing in series with the drain anyway? Does that actually do anything?
Looks like it does not do anything useful but causes some inefficiency.

Also, you might want to look into the drive circuit of the MOSFET. Using a 100 Ohm resistor in series with the gate means the gate peak current will be limited. That means the MOSFET will not turn on as fast as it could and not off as fast as it could. That also leads to inefficiencies.
Modern MOSFET drive circuits use dedicated MOSFET driver IC's but you can also use transistors which at least work a little better than resistors.

 

The operating frequency will be 150Hz.
If I remove the 100R at the Gate will that drive it fully on? I’m not overly concerned with efficiency here and an IC driver will add complication.

The IN5400 was part of the original design and I assume was put there for some form of protection of the FET.

The diode between Drain and Ground has been replaced with a 0.2uF, 600V ignition condenser to fulfil the same role that it does with the old style points in a distributor - to stop them burning out so it mops up charge when the FET switches.

 

TS's data: Vps=12 V; HV=10-15 kV; C=0.2 μF; Rload=1.6 MΩ; F=150 Hz.
-------------------------------------------------------------------------------

ADDED:
So, in this circuit transistor is fully protected:
Drain voltage in circuit is 190 V peak (allowed up to 400 V).
Drain current is 1.83 A peak (allowed up to10 A).
Body diode current is 0.264 A peak (allowed up to10A).
-----------------------------------------------------------------------
Power, dissipated by transistor is 90 mW. Power consumed from PS 12 V is 8 W. HV power is 5.5 W @ load 1.6 MΩ.

 

Why not use one of the IGBTs designed for ignition use?
https://www.onsemi.com/products/dis...1ZX4yfkFFQyBRVUFMSUZJRUR+UFBBUCBDQVBBQkxFfg==

 

Why not use one of the IGBTs designed for ignition use?
https://www.onsemi.com/products/dis...1ZX4yfkFFQyBRVUFMSUZJRUR+UFBBUCBDQVBBQkxFfg==

A 1200V, 40 A one? Would I still need the IN4500, or something similar, and the condenser in parallel with it to protect it?

 

A 1200V, 40 A one? Would I still need the IN4500, or something similar, and the condenser in parallel with it to protect it?

Not the 1N5400.
In theory not the capacitor, because there are no contacts to protect from sparks, but it may be contributing something useful in energy storage. I presume there are application notes. . .

 

Check saturation time for coil as it should be less than 1.5ms. Use a standard automotive condenser across either the coil or mosfet which should be rated at least 600V and 20A. Fully saturating the coil will only create substantial heating of coil and mosfet. FWIW I would use a 1KV rated SICFET instead but be very careful of the drive circuit.
Switching speed is high enough that the gate must be protected by a zener from going more than about 10V negative. That mistake cost me a mosfet and an IGBT before I dug deeply into motor drive gate protection circuits.

 

Check saturation time for coil as it should be less than 1.5ms. Use a standard automotive condenser across either the coil or mosfet which should be rated at least 600V and 20A. Fully saturating the coil will only create substantial heating of coil and mosfet. FWIW I would use a 1KV rated SICFET instead but be very careful of the drive circuit.
Switching speed is high enough that the gate must be protected by a zener from going more than about 10V negative. That mistake cost me a mosfet and an IGBT before I dug deeply into motor drive gate protection circuits.

Yes I acquired an ignition condenser of a suitable rating. So a 10V Zener should go between ground and the Gate with its Cathode towards the Gate, even with a FET? Can’t visualise it at the moment.

 

Put it in series with a 15V Zener with it's anode towards the gate. If the gate is not clamped within the gate to source voltage parameters gate to drain capacitance of only a few pf will drive gate to source voltage too far during switching. Found the info on a page about gate driving circuit designs and issues.