1
\$\begingroup\$

I am trying to design a circuit to drive an ignition coil in an automotive application using a microcontroller. The microcontroller operates on a 5VDC input and puts out a 3.3V and 4mA digital signal from its pins. Attached is a picture of the circuit I have so far.

There are a few issues I know I need to resolve:

  1. I don’t know what model of MOSFET to use here. I’m having trouble using the digikey search and filter functions to figure out what I need. I am pretty certain I need an n channel enhancement mode MOSFET, though.
  2. I don’t know if I need to add another MOSFET before the one I already have in the circuit to drive it. Since I’m only working with 4mA and I need to drive upwards of 10A to the coil, I feel like I may need another MOSFET in between to step up the power to drive the main power MOSFET.
  3. I don’t know if I need additional protections added into my circuit. I’ve been reading about flywheel or flyback diodes, adding a resistor and capacitor in parallel with the coil, and zener diodes. These are beyond what I know about electronics.

If you notice anything else wrong with my circuit please let me know. I don’t usually work with these kind of components; I’m used to just wiring everything up to the car battery.

Thanks, Zackcircuit draft

Edit: updated schematic 1: enter image description here

Edit 2: Is an IGBT what I need maybe? This looks like it could do what I want? Having trouble understanding if it meets the specs I need, though. I've never read anything about IGBTs until today.

\$\endgroup\$
6
  • 1
    \$\begingroup\$ I think you should consider using a MOSFET gate driver IC for this circuit to go between your micro and the big MOSFET driving the ignition coil. For power switching applications like this it's important to ensure fast turn-on and turn-off times - which is something you don't really want to bother with trying to achieve with extra discrete components when you could just drop in a single IC. \$\endgroup\$
    – brhans
    Mar 21 at 18:47
  • \$\begingroup\$ It might also benefit you to investigate using a 'snubber' circuit across the MOSFET instead of just the single flyback diode. \$\endgroup\$
    – brhans
    Mar 21 at 18:48
  • \$\begingroup\$ I think it would be better if you connect drain of first mosfet to the gate of second mosfet in your updated schematic. \$\endgroup\$
    – knight
    Mar 22 at 3:37
  • \$\begingroup\$ Okay, I have narrowed my choices for Ignition Specific IGBTs to these, but having trouble figuring out which meet my specs. Going to continue researching. Any input it appreciated. Also exploring the possibility of a gate driver IC and MOSFET combo like @brhans suggested. \$\endgroup\$
    – Zsn0w
    Mar 22 at 17:46
  • \$\begingroup\$ @knight If I were to do that, would I then need a p channel mosfet for the first one? \$\endgroup\$
    – Zsn0w
    Mar 22 at 17:46

3 Answers 3

Reset to default
2
\$\begingroup\$

With an ignition coil you need to let it flyback to 300-400V otherwise you won’t get much of a spark. As the OP has found, there are specific devices that are designed to allow this.

The other consideration is current limiting. With an ignition coil you apply current, then hold it until you want to fire - you turn the igbt off. Older ignition coils used with points are designed to not saturate whereas modern coils will and require the current to be limited.

There’s plenty of information and examples on the megasquirt site. You might want to consider a coil driver module that are common spare parts for cars or maybe a ignition module that incorporates the required circuitry and just needs a 5V signal to trigger. I’ve used Toyota Corolla coil on plug units that have this feature.

\$\endgroup\$
1
  • \$\begingroup\$ mouser.com/c/semiconductors/power-management-ics/… I'm thinking these are what I need to give me the flyback you're talking about? Having trouble understanding what exactly the voltage clamping they talk about is. Is that just the flyback limiting? Continuing to research all of these to try to decide which one of these to go with/if they'll accomplish what I need. Thanks for your response. \$\endgroup\$
    – Zsn0w
    Mar 22 at 18:18
1
\$\begingroup\$

  1. You need a beefy power MOSFET such as the IRF3205. But you need a gate-source voltage of higher than 3.3V to fully turn ON the power MOSFET. Check the datasheet for more details.

  2. Yes, add a bjt or mosfet and connect collector/drain to gate of power mosfet.

  3. Again, you need a power semiconductor component, a diode, between Supply and Drain. Anode of diode to drain and cathode to supply voltage.

\$\endgroup\$
12
  • \$\begingroup\$ Thank you so much for the answer. I’ll attach an updated schematic, let me know if that’s what you mean and if I have the diodes correct if you can. Just need to find a mosfet to use for the first part now. Think based on what I drew up it needs to be able to handle 10 V at .5 A to be safe. Maybe I should use bigger resistors on the divider going into the first MOSFET to the amperage down more? Voltage needs to be about 8 from the data sheet of the IRF3205 I think? Thanks again. \$\endgroup\$
    – Zsn0w
    Mar 21 at 15:59
  • \$\begingroup\$ I am thinking maybe 500 and 1000 ohms for the resistors in the voltage divider because that gives me between 7.7 and 9.7 volts in the normal car battery voltage range of 12-14.5 and limits the amps to under 30 milliamps. \$\endgroup\$
    – Zsn0w
    Mar 21 at 16:13
  • \$\begingroup\$ Cascading two MOSFETS like that means you need to add the Vgs values together, so your micro has no hope of turning them both fully on with a 3.3V output. You'd be far better using a Darlington NPN. \$\endgroup\$
    – Finbarr
    Mar 21 at 16:20
  • \$\begingroup\$ It can't just use the ~8 Volts I have coming in off of the 12-14V battery to turn the second MOSFET on? Why does it add the Vgs together? I had looked at BJTs but I was afraid they might not be fast enough, since I couldn't find any listed switching times in the datasheets for them like I could in MOSFET sheets and they are supposed to be slower. \$\endgroup\$
    – Zsn0w
    Mar 21 at 16:42
  • \$\begingroup\$ The source of the first connects to the gate of the second, so the Vgs of the first adds to the Vgs of the second. Though it's largely academic as you won't even be able to turn the IRF3205 on enough with just 3.3V \$\endgroup\$
    – Finbarr
    Mar 21 at 16:49
0
\$\begingroup\$

I've seen this done with an NPN bipolar transistor for the positive drive of the coil waveform and an PNP for the negative. You have to look at the coil pack to see whether it uses a positive or negative waveform to trigger. Note the coil received 12v on a separate input - the trigger signal is fine at 5V - for a 4 wire coil pack, which seem very popular.

enter image description here

\$\endgroup\$
2
  • \$\begingroup\$ I get the impression from the OP's question that it's an old fashioned "raw" ignition coil - basically just a very high-turns-ratio autotransformer - and if that's the case it needs to driven pretty hard by an open-collector or open-drain, and little 2N2222/3906 BJTs + 10R series resistor won't even begin to do the job. But that impression of mine could easily just be my own preconceptions kicking in because it's what I did 15-ish years ago :P. @Zsn0w can you clarify for us what kind of ignition coil you're using? \$\endgroup\$
    – brhans
    Mar 21 at 19:28
  • \$\begingroup\$ @brhans that is correct, I am working with an old school cylinder ignition coil. Specifically, I am thinking about using this Pertronix model pertronixbrands.com/products/…. I have also been looking at the e-core style coils, but it is hard to find information on the factory coils re: what circuitry they do/do not already have inside them and what their specs are. \$\endgroup\$
    – Zsn0w
    Mar 22 at 11:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.